diff --git a/Python/Differences_in_Differences.ipynb b/Python/Differences_in_Differences.ipynb
new file mode 100644
index 0000000..9c6cc6a
--- /dev/null
+++ b/Python/Differences_in_Differences.ipynb
@@ -0,0 +1,2843 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Difference-in-Differences** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "\n",
+    "import plotnine as p"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_stata(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "abortion = read_data('abortion.dta')\n",
+    "abortion = abortion[~pd.isnull(abortion.lnr)]\n",
+    "abortion_bf15 = abortion[abortion.bf15==1]\n",
+    "\n",
+    "reg = sm.OLS.from_formula(\"\"\"lnr ~ repeal*C(year) + C(fip) + acc + ir + pi + alcohol + crack + poverty + income + ur\"\"\",\n",
+    "            data = abortion_bf15, freq_weights=abortion_bf15['totpop']).fit(\n",
+    "            cov_type='cluster', cov_kwds={'groups': abortion_bf15['fip'].values})"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 89, but rank is 27\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>lnr</td>       <th>  R-squared:         </th> <td>   0.823</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.799</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   172.3</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>1.30e-40</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:30:26</td>     <th>  Log-Likelihood:    </th> <td> -193.93</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   737</td>      <th>  AIC:               </th> <td>   565.9</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   648</td>      <th>  BIC:               </th> <td>   975.5</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    88</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "              <td></td>                <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>                <td>    7.3038</td> <td>    0.726</td> <td>   10.062</td> <td> 0.000</td> <td>    5.881</td> <td>    8.726</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1986.0]</th>        <td>   -0.0229</td> <td>    0.058</td> <td>   -0.397</td> <td> 0.691</td> <td>   -0.136</td> <td>    0.090</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1987.0]</th>        <td>   -0.1612</td> <td>    0.070</td> <td>   -2.311</td> <td> 0.021</td> <td>   -0.298</td> <td>   -0.024</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1988.0]</th>        <td>   -0.0690</td> <td>    0.094</td> <td>   -0.733</td> <td> 0.463</td> <td>   -0.253</td> <td>    0.115</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1989.0]</th>        <td>   -0.1014</td> <td>    0.117</td> <td>   -0.867</td> <td> 0.386</td> <td>   -0.331</td> <td>    0.128</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1990.0]</th>        <td>   -0.2153</td> <td>    0.133</td> <td>   -1.618</td> <td> 0.106</td> <td>   -0.476</td> <td>    0.046</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1991.0]</th>        <td>   -0.2629</td> <td>    0.136</td> <td>   -1.940</td> <td> 0.052</td> <td>   -0.528</td> <td>    0.003</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1992.0]</th>        <td>   -0.4660</td> <td>    0.164</td> <td>   -2.847</td> <td> 0.004</td> <td>   -0.787</td> <td>   -0.145</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1993.0]</th>        <td>   -0.5735</td> <td>    0.176</td> <td>   -3.258</td> <td> 0.001</td> <td>   -0.919</td> <td>   -0.229</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1994.0]</th>        <td>   -0.6423</td> <td>    0.217</td> <td>   -2.956</td> <td> 0.003</td> <td>   -1.068</td> <td>   -0.216</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1995.0]</th>        <td>   -0.8025</td> <td>    0.234</td> <td>   -3.423</td> <td> 0.001</td> <td>   -1.262</td> <td>   -0.343</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1996.0]</th>        <td>   -1.1184</td> <td>    0.266</td> <td>   -4.197</td> <td> 0.000</td> <td>   -1.641</td> <td>   -0.596</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1997.0]</th>        <td>   -1.1328</td> <td>    0.280</td> <td>   -4.042</td> <td> 0.000</td> <td>   -1.682</td> <td>   -0.584</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1998.0]</th>        <td>   -1.0992</td> <td>    0.333</td> <td>   -3.300</td> <td> 0.001</td> <td>   -1.752</td> <td>   -0.446</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1999.0]</th>        <td>   -1.1324</td> <td>    0.347</td> <td>   -3.260</td> <td> 0.001</td> <td>   -1.813</td> <td>   -0.452</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2000.0]</th>        <td>   -1.3389</td> <td>    0.408</td> <td>   -3.281</td> <td> 0.001</td> <td>   -2.139</td> <td>   -0.539</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.2.0]</th>            <td>   -0.6991</td> <td>    0.197</td> <td>   -3.555</td> <td> 0.000</td> <td>   -1.084</td> <td>   -0.314</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.4.0]</th>            <td>   -0.5558</td> <td>    0.200</td> <td>   -2.780</td> <td> 0.005</td> <td>   -0.948</td> <td>   -0.164</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.5.0]</th>            <td>    0.1027</td> <td>    0.048</td> <td>    2.132</td> <td> 0.033</td> <td>    0.008</td> <td>    0.197</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.6.0]</th>            <td>    0.1596</td> <td>    0.074</td> <td>    2.163</td> <td> 0.031</td> <td>    0.015</td> <td>    0.304</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.8.0]</th>            <td>   -0.6853</td> <td>    0.263</td> <td>   -2.608</td> <td> 0.009</td> <td>   -1.200</td> <td>   -0.170</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.9.0]</th>            <td>   -1.0868</td> <td>    0.381</td> <td>   -2.849</td> <td> 0.004</td> <td>   -1.834</td> <td>   -0.339</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.10.0]</th>           <td>   -0.4074</td> <td>    0.305</td> <td>   -1.334</td> <td> 0.182</td> <td>   -1.006</td> <td>    0.191</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.11.0]</th>           <td>   -1.7034</td> <td>    0.652</td> <td>   -2.614</td> <td> 0.009</td> <td>   -2.981</td> <td>   -0.426</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.12.0]</th>           <td>   -0.8086</td> <td>    0.243</td> <td>   -3.325</td> <td> 0.001</td> <td>   -1.285</td> <td>   -0.332</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.13.0]</th>           <td>   -0.4840</td> <td>    0.124</td> <td>   -3.905</td> <td> 0.000</td> <td>   -0.727</td> <td>   -0.241</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.15.0]</th>           <td>   -0.5594</td> <td>    0.059</td> <td>   -9.468</td> <td> 0.000</td> <td>   -0.675</td> <td>   -0.444</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.16.0]</th>           <td>   -1.5289</td> <td>    0.109</td> <td>  -13.963</td> <td> 0.000</td> <td>   -1.744</td> <td>   -1.314</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.17.0]</th>           <td>   -0.5184</td> <td>    0.254</td> <td>   -2.044</td> <td> 0.041</td> <td>   -1.016</td> <td>   -0.021</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.18.0]</th>           <td>    0.0310</td> <td>    0.094</td> <td>    0.329</td> <td> 0.742</td> <td>   -0.153</td> <td>    0.215</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.19.0]</th>           <td>   -0.0302</td> <td>    0.090</td> <td>   -0.335</td> <td> 0.737</td> <td>   -0.206</td> <td>    0.146</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.20.0]</th>           <td>    0.2988</td> <td>    0.093</td> <td>    3.222</td> <td> 0.001</td> <td>    0.117</td> <td>    0.481</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.21.0]</th>           <td>    0.0706</td> <td>    0.082</td> <td>    0.859</td> <td> 0.390</td> <td>   -0.090</td> <td>    0.232</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.22.0]</th>           <td>   -0.6455</td> <td>    0.108</td> <td>   -5.974</td> <td> 0.000</td> <td>   -0.857</td> <td>   -0.434</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.23.0]</th>           <td>   -2.1869</td> <td>    0.130</td> <td>  -16.777</td> <td> 0.000</td> <td>   -2.442</td> <td>   -1.931</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.24.0]</th>           <td>   -1.0412</td> <td>    0.262</td> <td>   -3.980</td> <td> 0.000</td> <td>   -1.554</td> <td>   -0.528</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.25.0]</th>           <td>   -1.3159</td> <td>    0.330</td> <td>   -3.983</td> <td> 0.000</td> <td>   -1.963</td> <td>   -0.668</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.26.0]</th>           <td>   -0.6024</td> <td>    0.172</td> <td>   -3.496</td> <td> 0.000</td> <td>   -0.940</td> <td>   -0.265</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.27.0]</th>           <td>   -0.1472</td> <td>    0.212</td> <td>   -0.696</td> <td> 0.487</td> <td>   -0.562</td> <td>    0.267</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.28.0]</th>           <td>   -0.0821</td> <td>    0.086</td> <td>   -0.949</td> <td> 0.343</td> <td>   -0.252</td> <td>    0.087</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.29.0]</th>           <td>    0.0632</td> <td>    0.120</td> <td>    0.526</td> <td> 0.599</td> <td>   -0.172</td> <td>    0.299</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.30.0]</th>           <td>   -1.3616</td> <td>    0.182</td> <td>   -7.485</td> <td> 0.000</td> <td>   -1.718</td> <td>   -1.005</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.31.0]</th>           <td>   -0.0058</td> <td>    0.135</td> <td>   -0.043</td> <td> 0.966</td> <td>   -0.270</td> <td>    0.258</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.32.0]</th>           <td>   -1.6028</td> <td>    0.552</td> <td>   -2.904</td> <td> 0.004</td> <td>   -2.685</td> <td>   -0.521</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.33.0]</th>           <td>   -3.5197</td> <td>    0.561</td> <td>   -6.271</td> <td> 0.000</td> <td>   -4.620</td> <td>   -2.420</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.34.0]</th>           <td>   -1.7814</td> <td>    0.338</td> <td>   -5.271</td> <td> 0.000</td> <td>   -2.444</td> <td>   -1.119</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.35.0]</th>           <td>   -0.8645</td> <td>    0.159</td> <td>   -5.452</td> <td> 0.000</td> <td>   -1.175</td> <td>   -0.554</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.36.0]</th>           <td>   -0.7651</td> <td>    0.077</td> <td>   -9.924</td> <td> 0.000</td> <td>   -0.916</td> <td>   -0.614</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.37.0]</th>           <td>    0.0440</td> <td>    0.083</td> <td>    0.529</td> <td> 0.597</td> <td>   -0.119</td> <td>    0.207</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.38.0]</th>           <td>   -2.0122</td> <td>    0.130</td> <td>  -15.487</td> <td> 0.000</td> <td>   -2.267</td> <td>   -1.758</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.39.0]</th>           <td>   -0.1986</td> <td>    0.128</td> <td>   -1.547</td> <td> 0.122</td> <td>   -0.450</td> <td>    0.053</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.40.0]</th>           <td>    0.4134</td> <td>    0.071</td> <td>    5.783</td> <td> 0.000</td> <td>    0.273</td> <td>    0.554</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.41.0]</th>           <td>   -0.6701</td> <td>    0.210</td> <td>   -3.189</td> <td> 0.001</td> <td>   -1.082</td> <td>   -0.258</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.42.0]</th>           <td>   -0.1478</td> <td>    0.171</td> <td>   -0.864</td> <td> 0.387</td> <td>   -0.483</td> <td>    0.187</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.44.0]</th>           <td>   -0.5483</td> <td>    0.229</td> <td>   -2.393</td> <td> 0.017</td> <td>   -0.997</td> <td>   -0.099</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.45.0]</th>           <td>   -1.0164</td> <td>    0.105</td> <td>   -9.716</td> <td> 0.000</td> <td>   -1.221</td> <td>   -0.811</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.46.0]</th>           <td>   -1.3652</td> <td>    0.142</td> <td>   -9.582</td> <td> 0.000</td> <td>   -1.644</td> <td>   -1.086</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.47.0]</th>           <td>    0.1857</td> <td>    0.073</td> <td>    2.529</td> <td> 0.011</td> <td>    0.042</td> <td>    0.330</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.48.0]</th>           <td>   -0.4381</td> <td>    0.160</td> <td>   -2.742</td> <td> 0.006</td> <td>   -0.751</td> <td>   -0.125</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.49.0]</th>           <td>   -0.5898</td> <td>    0.224</td> <td>   -2.638</td> <td> 0.008</td> <td>   -1.028</td> <td>   -0.152</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.50.0]</th>           <td>   -1.9428</td> <td>    0.224</td> <td>   -8.670</td> <td> 0.000</td> <td>   -2.382</td> <td>   -1.504</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.51.0]</th>           <td>   -0.6569</td> <td>    0.180</td> <td>   -3.659</td> <td> 0.000</td> <td>   -1.009</td> <td>   -0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.53.0]</th>           <td>    0.5970</td> <td>    0.072</td> <td>    8.323</td> <td> 0.000</td> <td>    0.456</td> <td>    0.738</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.54.0]</th>           <td>   -0.4510</td> <td>    0.101</td> <td>   -4.455</td> <td> 0.000</td> <td>   -0.649</td> <td>   -0.253</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.55.0]</th>           <td>   -0.1693</td> <td>    0.229</td> <td>   -0.738</td> <td> 0.461</td> <td>   -0.619</td> <td>    0.280</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.56.0]</th>           <td>   -1.8681</td> <td>    0.164</td> <td>  -11.369</td> <td> 0.000</td> <td>   -2.190</td> <td>   -1.546</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal</th>                   <td>   -1.2671</td> <td>    0.282</td> <td>   -4.499</td> <td> 0.000</td> <td>   -1.819</td> <td>   -0.715</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1986.0]</th> <td>   -0.0051</td> <td>    0.186</td> <td>   -0.027</td> <td> 0.978</td> <td>   -0.369</td> <td>    0.359</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1987.0]</th> <td>   -0.0279</td> <td>    0.190</td> <td>   -0.146</td> <td> 0.884</td> <td>   -0.401</td> <td>    0.345</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1988.0]</th> <td>   -0.2724</td> <td>    0.322</td> <td>   -0.845</td> <td> 0.398</td> <td>   -0.904</td> <td>    0.359</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1989.0]</th> <td>   -0.2384</td> <td>    0.309</td> <td>   -0.772</td> <td> 0.440</td> <td>   -0.844</td> <td>    0.367</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1990.0]</th> <td>   -0.1571</td> <td>    0.339</td> <td>   -0.463</td> <td> 0.643</td> <td>   -0.822</td> <td>    0.508</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1991.0]</th> <td>   -0.0898</td> <td>    0.186</td> <td>   -0.484</td> <td> 0.629</td> <td>   -0.454</td> <td>    0.274</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1992.0]</th> <td>    0.0235</td> <td>    0.242</td> <td>    0.097</td> <td> 0.923</td> <td>   -0.452</td> <td>    0.499</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1993.0]</th> <td>    0.1969</td> <td>    0.356</td> <td>    0.553</td> <td> 0.580</td> <td>   -0.501</td> <td>    0.895</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1994.0]</th> <td>    0.2779</td> <td>    0.354</td> <td>    0.785</td> <td> 0.433</td> <td>   -0.416</td> <td>    0.972</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1995.0]</th> <td>    0.1522</td> <td>    0.336</td> <td>    0.453</td> <td> 0.651</td> <td>   -0.507</td> <td>    0.811</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1996.0]</th> <td>   -0.1404</td> <td>    0.239</td> <td>   -0.587</td> <td> 0.557</td> <td>   -0.609</td> <td>    0.329</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1997.0]</th> <td>    0.0698</td> <td>    0.250</td> <td>    0.279</td> <td> 0.780</td> <td>   -0.421</td> <td>    0.561</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1998.0]</th> <td>   -0.3285</td> <td>    0.284</td> <td>   -1.155</td> <td> 0.248</td> <td>   -0.886</td> <td>    0.229</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.1999.0]</th> <td>   -0.2417</td> <td>    0.333</td> <td>   -0.726</td> <td> 0.468</td> <td>   -0.894</td> <td>    0.411</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>repeal:C(year)[T.2000.0]</th> <td>   -0.0824</td> <td>    0.290</td> <td>   -0.284</td> <td> 0.776</td> <td>   -0.650</td> <td>    0.485</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>acc</th>                      <td>    0.0016</td> <td>    0.001</td> <td>    1.833</td> <td> 0.067</td> <td>   -0.000</td> <td>    0.003</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ir</th>                       <td>-1.754e-05</td> <td> 9.34e-05</td> <td>   -0.188</td> <td> 0.851</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>pi</th>                       <td>   -0.0157</td> <td>    0.078</td> <td>   -0.202</td> <td> 0.840</td> <td>   -0.168</td> <td>    0.137</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>alcohol</th>                  <td>    0.4583</td> <td>    0.187</td> <td>    2.454</td> <td> 0.014</td> <td>    0.092</td> <td>    0.824</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>crack</th>                    <td>   -0.0662</td> <td>    0.044</td> <td>   -1.489</td> <td> 0.137</td> <td>   -0.153</td> <td>    0.021</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>poverty</th>                  <td>   -0.0049</td> <td>    0.010</td> <td>   -0.467</td> <td> 0.640</td> <td>   -0.025</td> <td>    0.015</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>income</th>                   <td>  5.27e-05</td> <td> 3.04e-05</td> <td>    1.734</td> <td> 0.083</td> <td>-6.86e-06</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ur</th>                       <td>    0.0004</td> <td>    0.025</td> <td>    0.016</td> <td> 0.987</td> <td>   -0.048</td> <td>    0.049</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>124.661</td> <th>  Durbin-Watson:     </th> <td>   1.102</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td>  <th>  Jarque-Bera (JB):  </th> <td> 604.691</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.671</td>  <th>  Prob(JB):          </th> <td>4.93e-132</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 7.230</td>  <th>  Cond. No.          </th> <td>1.93e+20</td> \n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 9.25e-30. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                    lnr   R-squared:                       0.823\n",
+       "Model:                            OLS   Adj. R-squared:                  0.799\n",
+       "Method:                 Least Squares   F-statistic:                     172.3\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           1.30e-40\n",
+       "Time:                        13:30:26   Log-Likelihood:                -193.93\n",
+       "No. Observations:                 737   AIC:                             565.9\n",
+       "Df Residuals:                     648   BIC:                             975.5\n",
+       "Df Model:                          88                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "============================================================================================\n",
+       "                               coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "--------------------------------------------------------------------------------------------\n",
+       "Intercept                    7.3038      0.726     10.062      0.000       5.881       8.726\n",
+       "C(year)[T.1986.0]           -0.0229      0.058     -0.397      0.691      -0.136       0.090\n",
+       "C(year)[T.1987.0]           -0.1612      0.070     -2.311      0.021      -0.298      -0.024\n",
+       "C(year)[T.1988.0]           -0.0690      0.094     -0.733      0.463      -0.253       0.115\n",
+       "C(year)[T.1989.0]           -0.1014      0.117     -0.867      0.386      -0.331       0.128\n",
+       "C(year)[T.1990.0]           -0.2153      0.133     -1.618      0.106      -0.476       0.046\n",
+       "C(year)[T.1991.0]           -0.2629      0.136     -1.940      0.052      -0.528       0.003\n",
+       "C(year)[T.1992.0]           -0.4660      0.164     -2.847      0.004      -0.787      -0.145\n",
+       "C(year)[T.1993.0]           -0.5735      0.176     -3.258      0.001      -0.919      -0.229\n",
+       "C(year)[T.1994.0]           -0.6423      0.217     -2.956      0.003      -1.068      -0.216\n",
+       "C(year)[T.1995.0]           -0.8025      0.234     -3.423      0.001      -1.262      -0.343\n",
+       "C(year)[T.1996.0]           -1.1184      0.266     -4.197      0.000      -1.641      -0.596\n",
+       "C(year)[T.1997.0]           -1.1328      0.280     -4.042      0.000      -1.682      -0.584\n",
+       "C(year)[T.1998.0]           -1.0992      0.333     -3.300      0.001      -1.752      -0.446\n",
+       "C(year)[T.1999.0]           -1.1324      0.347     -3.260      0.001      -1.813      -0.452\n",
+       "C(year)[T.2000.0]           -1.3389      0.408     -3.281      0.001      -2.139      -0.539\n",
+       "C(fip)[T.2.0]               -0.6991      0.197     -3.555      0.000      -1.084      -0.314\n",
+       "C(fip)[T.4.0]               -0.5558      0.200     -2.780      0.005      -0.948      -0.164\n",
+       "C(fip)[T.5.0]                0.1027      0.048      2.132      0.033       0.008       0.197\n",
+       "C(fip)[T.6.0]                0.1596      0.074      2.163      0.031       0.015       0.304\n",
+       "C(fip)[T.8.0]               -0.6853      0.263     -2.608      0.009      -1.200      -0.170\n",
+       "C(fip)[T.9.0]               -1.0868      0.381     -2.849      0.004      -1.834      -0.339\n",
+       "C(fip)[T.10.0]              -0.4074      0.305     -1.334      0.182      -1.006       0.191\n",
+       "C(fip)[T.11.0]              -1.7034      0.652     -2.614      0.009      -2.981      -0.426\n",
+       "C(fip)[T.12.0]              -0.8086      0.243     -3.325      0.001      -1.285      -0.332\n",
+       "C(fip)[T.13.0]              -0.4840      0.124     -3.905      0.000      -0.727      -0.241\n",
+       "C(fip)[T.15.0]              -0.5594      0.059     -9.468      0.000      -0.675      -0.444\n",
+       "C(fip)[T.16.0]              -1.5289      0.109    -13.963      0.000      -1.744      -1.314\n",
+       "C(fip)[T.17.0]              -0.5184      0.254     -2.044      0.041      -1.016      -0.021\n",
+       "C(fip)[T.18.0]               0.0310      0.094      0.329      0.742      -0.153       0.215\n",
+       "C(fip)[T.19.0]              -0.0302      0.090     -0.335      0.737      -0.206       0.146\n",
+       "C(fip)[T.20.0]               0.2988      0.093      3.222      0.001       0.117       0.481\n",
+       "C(fip)[T.21.0]               0.0706      0.082      0.859      0.390      -0.090       0.232\n",
+       "C(fip)[T.22.0]              -0.6455      0.108     -5.974      0.000      -0.857      -0.434\n",
+       "C(fip)[T.23.0]              -2.1869      0.130    -16.777      0.000      -2.442      -1.931\n",
+       "C(fip)[T.24.0]              -1.0412      0.262     -3.980      0.000      -1.554      -0.528\n",
+       "C(fip)[T.25.0]              -1.3159      0.330     -3.983      0.000      -1.963      -0.668\n",
+       "C(fip)[T.26.0]              -0.6024      0.172     -3.496      0.000      -0.940      -0.265\n",
+       "C(fip)[T.27.0]              -0.1472      0.212     -0.696      0.487      -0.562       0.267\n",
+       "C(fip)[T.28.0]              -0.0821      0.086     -0.949      0.343      -0.252       0.087\n",
+       "C(fip)[T.29.0]               0.0632      0.120      0.526      0.599      -0.172       0.299\n",
+       "C(fip)[T.30.0]              -1.3616      0.182     -7.485      0.000      -1.718      -1.005\n",
+       "C(fip)[T.31.0]              -0.0058      0.135     -0.043      0.966      -0.270       0.258\n",
+       "C(fip)[T.32.0]              -1.6028      0.552     -2.904      0.004      -2.685      -0.521\n",
+       "C(fip)[T.33.0]              -3.5197      0.561     -6.271      0.000      -4.620      -2.420\n",
+       "C(fip)[T.34.0]              -1.7814      0.338     -5.271      0.000      -2.444      -1.119\n",
+       "C(fip)[T.35.0]              -0.8645      0.159     -5.452      0.000      -1.175      -0.554\n",
+       "C(fip)[T.36.0]              -0.7651      0.077     -9.924      0.000      -0.916      -0.614\n",
+       "C(fip)[T.37.0]               0.0440      0.083      0.529      0.597      -0.119       0.207\n",
+       "C(fip)[T.38.0]              -2.0122      0.130    -15.487      0.000      -2.267      -1.758\n",
+       "C(fip)[T.39.0]              -0.1986      0.128     -1.547      0.122      -0.450       0.053\n",
+       "C(fip)[T.40.0]               0.4134      0.071      5.783      0.000       0.273       0.554\n",
+       "C(fip)[T.41.0]              -0.6701      0.210     -3.189      0.001      -1.082      -0.258\n",
+       "C(fip)[T.42.0]              -0.1478      0.171     -0.864      0.387      -0.483       0.187\n",
+       "C(fip)[T.44.0]              -0.5483      0.229     -2.393      0.017      -0.997      -0.099\n",
+       "C(fip)[T.45.0]              -1.0164      0.105     -9.716      0.000      -1.221      -0.811\n",
+       "C(fip)[T.46.0]              -1.3652      0.142     -9.582      0.000      -1.644      -1.086\n",
+       "C(fip)[T.47.0]               0.1857      0.073      2.529      0.011       0.042       0.330\n",
+       "C(fip)[T.48.0]              -0.4381      0.160     -2.742      0.006      -0.751      -0.125\n",
+       "C(fip)[T.49.0]              -0.5898      0.224     -2.638      0.008      -1.028      -0.152\n",
+       "C(fip)[T.50.0]              -1.9428      0.224     -8.670      0.000      -2.382      -1.504\n",
+       "C(fip)[T.51.0]              -0.6569      0.180     -3.659      0.000      -1.009      -0.305\n",
+       "C(fip)[T.53.0]               0.5970      0.072      8.323      0.000       0.456       0.738\n",
+       "C(fip)[T.54.0]              -0.4510      0.101     -4.455      0.000      -0.649      -0.253\n",
+       "C(fip)[T.55.0]              -0.1693      0.229     -0.738      0.461      -0.619       0.280\n",
+       "C(fip)[T.56.0]              -1.8681      0.164    -11.369      0.000      -2.190      -1.546\n",
+       "repeal                      -1.2671      0.282     -4.499      0.000      -1.819      -0.715\n",
+       "repeal:C(year)[T.1986.0]    -0.0051      0.186     -0.027      0.978      -0.369       0.359\n",
+       "repeal:C(year)[T.1987.0]    -0.0279      0.190     -0.146      0.884      -0.401       0.345\n",
+       "repeal:C(year)[T.1988.0]    -0.2724      0.322     -0.845      0.398      -0.904       0.359\n",
+       "repeal:C(year)[T.1989.0]    -0.2384      0.309     -0.772      0.440      -0.844       0.367\n",
+       "repeal:C(year)[T.1990.0]    -0.1571      0.339     -0.463      0.643      -0.822       0.508\n",
+       "repeal:C(year)[T.1991.0]    -0.0898      0.186     -0.484      0.629      -0.454       0.274\n",
+       "repeal:C(year)[T.1992.0]     0.0235      0.242      0.097      0.923      -0.452       0.499\n",
+       "repeal:C(year)[T.1993.0]     0.1969      0.356      0.553      0.580      -0.501       0.895\n",
+       "repeal:C(year)[T.1994.0]     0.2779      0.354      0.785      0.433      -0.416       0.972\n",
+       "repeal:C(year)[T.1995.0]     0.1522      0.336      0.453      0.651      -0.507       0.811\n",
+       "repeal:C(year)[T.1996.0]    -0.1404      0.239     -0.587      0.557      -0.609       0.329\n",
+       "repeal:C(year)[T.1997.0]     0.0698      0.250      0.279      0.780      -0.421       0.561\n",
+       "repeal:C(year)[T.1998.0]    -0.3285      0.284     -1.155      0.248      -0.886       0.229\n",
+       "repeal:C(year)[T.1999.0]    -0.2417      0.333     -0.726      0.468      -0.894       0.411\n",
+       "repeal:C(year)[T.2000.0]    -0.0824      0.290     -0.284      0.776      -0.650       0.485\n",
+       "acc                          0.0016      0.001      1.833      0.067      -0.000       0.003\n",
+       "ir                       -1.754e-05   9.34e-05     -0.188      0.851      -0.000       0.000\n",
+       "pi                          -0.0157      0.078     -0.202      0.840      -0.168       0.137\n",
+       "alcohol                      0.4583      0.187      2.454      0.014       0.092       0.824\n",
+       "crack                       -0.0662      0.044     -1.489      0.137      -0.153       0.021\n",
+       "poverty                     -0.0049      0.010     -0.467      0.640      -0.025       0.015\n",
+       "income                     5.27e-05   3.04e-05      1.734      0.083   -6.86e-06       0.000\n",
+       "ur                           0.0004      0.025      0.016      0.987      -0.048       0.049\n",
+       "==============================================================================\n",
+       "Omnibus:                      124.661   Durbin-Watson:                   1.102\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              604.691\n",
+       "Skew:                          -0.671   Prob(JB):                    4.93e-132\n",
+       "Kurtosis:                       7.230   Cond. No.                     1.93e+20\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 9.25e-30. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "reg.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8768291016683)>"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "abortion_plot = pd.DataFrame({'sd': reg.bse[2:17],\n",
+    "             'mean': reg.params.values[2:17],\n",
+    "             'year':np.arange(1986, 2001)})\n",
+    "abortion_plot['lb'] = abortion_plot['mean'] - abortion_plot['sd']*1.96\n",
+    "abortion_plot['ub'] = abortion_plot['mean'] + abortion_plot['sd']*1.96\n",
+    "\n",
+    "p.ggplot(abortion_plot, p.aes(x = 'year', y = 'mean')) + \\\n",
+    "    p.geom_rect(p.aes(xmin=1985, xmax=1992, ymin=-np.inf, ymax=np.inf), fill = \"cyan\", alpha = 0.01)+\\\n",
+    "    p.geom_point()+\\\n",
+    "    p.geom_text(p.aes(label = 'year'), ha='right')+\\\n",
+    "    p.geom_hline(yintercept = 0) +\\\n",
+    "    p.geom_errorbar(p.aes(ymin = 'lb', ymax = 'ub'), width = 0.2,\n",
+    "                position = p.position_dodge(0.05)) +\\\n",
+    "    p.labs(title= \"Estimated effect of abortion legalization on gonorrhea\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Note on results\n",
+    "\n",
+    "Python and R implementations produce different answers. Design matrix is rank deficient, `lm and lm_robust` have a convergence issue. `statsmodels` is more robust to rank deficiency."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Describe in your own words the testable predictions from the roll out of repeal versus Roe in the population?  In other words, describe the behavior of the DD coefficients under this regression.  \n",
+    "- Do we find evidence consistent with this in our DD analysis?  List all the evidence for and against the hypothesis in this analysis. \n",
+    "- Does it appear that there was an effect in the period where Roe has not fully caught up?\n",
+    "\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8768290925352)>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "abortion = read_data('abortion.dta')\n",
+    "abortion = abortion[~pd.isnull(abortion.lnr)]\n",
+    "\n",
+    "abortion['yr'] = 0\n",
+    "abortion.loc[(abortion.younger==1) & (abortion.repeal==1), 'yr'] = 1\n",
+    "\n",
+    "abortion['wm'] = 0\n",
+    "abortion.loc[(abortion.wht==1) & (abortion.male==1), 'wm'] = 1\n",
+    "\n",
+    "abortion['wf'] = 0\n",
+    "abortion.loc[(abortion.wht==1) & (abortion.male==0), 'wf'] = 1\n",
+    "\n",
+    "abortion['bm'] = 0\n",
+    "abortion.loc[(abortion.wht==0) & (abortion.male==1), 'bm'] = 1\n",
+    "\n",
+    "abortion['bf'] = 0\n",
+    "abortion.loc[(abortion.wht==0) & (abortion.male==0), 'bf'] = 1\n",
+    "\n",
+    "\n",
+    "abortion_filt = abortion[(abortion.bf==1) & (abortion.age.isin([15,25]))]\n",
+    "\n",
+    "reg = sm.OLS.from_formula(\"\"\"lnr ~ C(repeal)*C(year) + C(younger)*C(repeal) + C(younger)*C(year) + \n",
+    "C(yr)*C(year) + C(fip)*t + acc + ir + pi + alcohol + crack + poverty + income + ur\"\"\",\n",
+    "            data = abortion_filt, freq_weights = abortion_filt['totpop']).fit(\n",
+    "            cov_type='cluster', cov_kwds={'groups': abortion_filt['fip']})\n",
+    "\n",
+    "abortion_plot = pd.DataFrame({'sd': reg.bse[2:17],\n",
+    "             'mean': reg.params.values[2:17],\n",
+    "             'year':np.arange(1986, 2001)})\n",
+    "abortion_plot['lb'] = abortion_plot['mean'] - abortion_plot['sd']*1.96\n",
+    "abortion_plot['ub'] = abortion_plot['mean'] + abortion_plot['sd']*1.96\n",
+    "\n",
+    "p.ggplot(abortion_plot, p.aes(x = 'year', y = 'mean')) + \\\n",
+    "    p.geom_rect(p.aes(xmin=1986, xmax=1991, ymin=-np.inf, ymax=np.inf), fill = \"cyan\", alpha = 0.01)+\\\n",
+    "    p.geom_point()+\\\n",
+    "    p.geom_text(p.aes(label = 'year'), ha='right')+\\\n",
+    "    p.geom_hline(yintercept = 0) +\\\n",
+    "    p.geom_errorbar(p.aes(ymin = 'lb', ymax = 'ub'), width = 0.2,\n",
+    "                position = p.position_dodge(0.05)) +\\\n",
+    "    p.labs(title= \"Estimated effect of abortion legalization on gonorrhea\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Note on results\n",
+    "\n",
+    "Python and R implementations produce different answers. Design matrix is rank deficient, `lm and lm_robust` have a convergence issue. `statsmodels` is more robust to rank deficiency."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Why did we implement a triple difference?  What problems does this solve and to what degree do you feel it is a necessary check?\n",
+    "- Describe the evidence for and against the abortion selection hypothesis when using triple difference?  How is it consistent with our DD and how is it not?\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 89, but rank is 26\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>lnr</td>       <th>  R-squared:         </th> <td>   0.839</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.817</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   201.1</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>4.30e-42</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:30:28</td>     <th>  Log-Likelihood:    </th> <td> -191.07</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   733</td>      <th>  AIC:               </th> <td>   558.1</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   645</td>      <th>  BIC:               </th> <td>   962.7</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    87</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "                   <td></td>                     <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>                          <td>    6.6023</td> <td>    0.631</td> <td>   10.456</td> <td> 0.000</td> <td>    5.365</td> <td>    7.840</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]</th>                   <td>   -1.6162</td> <td>    0.278</td> <td>   -5.818</td> <td> 0.000</td> <td>   -2.161</td> <td>   -1.072</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1986.0]</th>                  <td>   -0.0142</td> <td>    0.055</td> <td>   -0.256</td> <td> 0.798</td> <td>   -0.122</td> <td>    0.094</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1987.0]</th>                  <td>   -0.1737</td> <td>    0.075</td> <td>   -2.327</td> <td> 0.020</td> <td>   -0.320</td> <td>   -0.027</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1988.0]</th>                  <td>   -0.1441</td> <td>    0.095</td> <td>   -1.513</td> <td> 0.130</td> <td>   -0.331</td> <td>    0.043</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1989.0]</th>                  <td>   -0.1632</td> <td>    0.134</td> <td>   -1.218</td> <td> 0.223</td> <td>   -0.426</td> <td>    0.099</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1990.0]</th>                  <td>   -0.3172</td> <td>    0.150</td> <td>   -2.109</td> <td> 0.035</td> <td>   -0.612</td> <td>   -0.022</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1991.0]</th>                  <td>   -0.3716</td> <td>    0.165</td> <td>   -2.246</td> <td> 0.025</td> <td>   -0.696</td> <td>   -0.047</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1992.0]</th>                  <td>   -0.5930</td> <td>    0.168</td> <td>   -3.539</td> <td> 0.000</td> <td>   -0.921</td> <td>   -0.265</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1993.0]</th>                  <td>   -0.7824</td> <td>    0.195</td> <td>   -4.007</td> <td> 0.000</td> <td>   -1.165</td> <td>   -0.400</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1994.0]</th>                  <td>   -0.7746</td> <td>    0.225</td> <td>   -3.439</td> <td> 0.001</td> <td>   -1.216</td> <td>   -0.333</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1995.0]</th>                  <td>   -0.9350</td> <td>    0.242</td> <td>   -3.856</td> <td> 0.000</td> <td>   -1.410</td> <td>   -0.460</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1996.0]</th>                  <td>   -1.1844</td> <td>    0.280</td> <td>   -4.228</td> <td> 0.000</td> <td>   -1.733</td> <td>   -0.635</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1997.0]</th>                  <td>   -1.2330</td> <td>    0.314</td> <td>   -3.925</td> <td> 0.000</td> <td>   -1.849</td> <td>   -0.617</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1998.0]</th>                  <td>   -1.1723</td> <td>    0.336</td> <td>   -3.491</td> <td> 0.000</td> <td>   -1.831</td> <td>   -0.514</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1999.0]</th>                  <td>   -1.2225</td> <td>    0.389</td> <td>   -3.143</td> <td> 0.002</td> <td>   -1.985</td> <td>   -0.460</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2000.0]</th>                  <td>   -1.4156</td> <td>    0.446</td> <td>   -3.177</td> <td> 0.001</td> <td>   -2.289</td> <td>   -0.542</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.2.0]</th>                      <td>   -0.8977</td> <td>    0.194</td> <td>   -4.635</td> <td> 0.000</td> <td>   -1.277</td> <td>   -0.518</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.4.0]</th>                      <td>   -0.9035</td> <td>    0.168</td> <td>   -5.381</td> <td> 0.000</td> <td>   -1.233</td> <td>   -0.574</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.5.0]</th>                      <td>    0.0064</td> <td>    0.045</td> <td>    0.144</td> <td> 0.886</td> <td>   -0.081</td> <td>    0.094</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.6.0]</th>                      <td>    0.0840</td> <td>    0.111</td> <td>    0.755</td> <td> 0.450</td> <td>   -0.134</td> <td>    0.302</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.8.0]</th>                      <td>   -0.9849</td> <td>    0.241</td> <td>   -4.080</td> <td> 0.000</td> <td>   -1.458</td> <td>   -0.512</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.9.0]</th>                      <td>   -1.3984</td> <td>    0.382</td> <td>   -3.664</td> <td> 0.000</td> <td>   -2.146</td> <td>   -0.650</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.10.0]</th>                     <td>   -0.9297</td> <td>    0.285</td> <td>   -3.261</td> <td> 0.001</td> <td>   -1.488</td> <td>   -0.371</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.11.0]</th>                     <td>   -2.2143</td> <td>    0.583</td> <td>   -3.795</td> <td> 0.000</td> <td>   -3.358</td> <td>   -1.071</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.12.0]</th>                     <td>   -0.9462</td> <td>    0.225</td> <td>   -4.211</td> <td> 0.000</td> <td>   -1.387</td> <td>   -0.506</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.13.0]</th>                     <td>   -0.4691</td> <td>    0.118</td> <td>   -3.984</td> <td> 0.000</td> <td>   -0.700</td> <td>   -0.238</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.15.0]</th>                     <td>   -0.7590</td> <td>    0.097</td> <td>   -7.812</td> <td> 0.000</td> <td>   -0.949</td> <td>   -0.569</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.16.0]</th>                     <td>   -1.5459</td> <td>    0.092</td> <td>  -16.826</td> <td> 0.000</td> <td>   -1.726</td> <td>   -1.366</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.17.0]</th>                     <td>   -0.7829</td> <td>    0.233</td> <td>   -3.363</td> <td> 0.001</td> <td>   -1.239</td> <td>   -0.327</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.18.0]</th>                     <td>   -0.2189</td> <td>    0.114</td> <td>   -1.920</td> <td> 0.055</td> <td>   -0.442</td> <td>    0.005</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.19.0]</th>                     <td>   -0.3851</td> <td>    0.204</td> <td>   -1.887</td> <td> 0.059</td> <td>   -0.785</td> <td>    0.015</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.20.0]</th>                     <td>    0.1033</td> <td>    0.119</td> <td>    0.870</td> <td> 0.384</td> <td>   -0.129</td> <td>    0.336</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.21.0]</th>                     <td>   -0.1655</td> <td>    0.049</td> <td>   -3.396</td> <td> 0.001</td> <td>   -0.261</td> <td>   -0.070</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.22.0]</th>                     <td>   -0.8190</td> <td>    0.107</td> <td>   -7.626</td> <td> 0.000</td> <td>   -1.030</td> <td>   -0.609</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.23.0]</th>                     <td>   -2.2270</td> <td>    0.140</td> <td>  -15.852</td> <td> 0.000</td> <td>   -2.502</td> <td>   -1.952</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.24.0]</th>                     <td>   -1.3288</td> <td>    0.260</td> <td>   -5.110</td> <td> 0.000</td> <td>   -1.839</td> <td>   -0.819</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.25.0]</th>                     <td>   -1.5452</td> <td>    0.312</td> <td>   -4.951</td> <td> 0.000</td> <td>   -2.157</td> <td>   -0.933</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.26.0]</th>                     <td>   -0.9182</td> <td>    0.178</td> <td>   -5.158</td> <td> 0.000</td> <td>   -1.267</td> <td>   -0.569</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.27.0]</th>                     <td>   -0.7059</td> <td>    0.235</td> <td>   -3.002</td> <td> 0.003</td> <td>   -1.167</td> <td>   -0.245</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.28.0]</th>                     <td>   -0.0630</td> <td>    0.084</td> <td>   -0.752</td> <td> 0.452</td> <td>   -0.227</td> <td>    0.101</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.29.0]</th>                     <td>   -0.1710</td> <td>    0.123</td> <td>   -1.388</td> <td> 0.165</td> <td>   -0.413</td> <td>    0.071</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.30.0]</th>                     <td>   -1.5919</td> <td>    0.158</td> <td>  -10.102</td> <td> 0.000</td> <td>   -1.901</td> <td>   -1.283</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.31.0]</th>                     <td>   -0.2417</td> <td>    0.149</td> <td>   -1.623</td> <td> 0.105</td> <td>   -0.534</td> <td>    0.050</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.32.0]</th>                     <td>   -2.4190</td> <td>    0.499</td> <td>   -4.848</td> <td> 0.000</td> <td>   -3.397</td> <td>   -1.441</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.33.0]</th>                     <td>   -3.9837</td> <td>    0.533</td> <td>   -7.474</td> <td> 0.000</td> <td>   -5.028</td> <td>   -2.939</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.34.0]</th>                     <td>   -2.1379</td> <td>    0.334</td> <td>   -6.408</td> <td> 0.000</td> <td>   -2.792</td> <td>   -1.484</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.35.0]</th>                     <td>   -1.1200</td> <td>    0.107</td> <td>  -10.453</td> <td> 0.000</td> <td>   -1.330</td> <td>   -0.910</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.36.0]</th>                     <td>   -0.5342</td> <td>    0.104</td> <td>   -5.149</td> <td> 0.000</td> <td>   -0.738</td> <td>   -0.331</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.37.0]</th>                     <td>   -0.0919</td> <td>    0.084</td> <td>   -1.092</td> <td> 0.275</td> <td>   -0.257</td> <td>    0.073</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.38.0]</th>                     <td>   -2.2710</td> <td>    0.133</td> <td>  -17.115</td> <td> 0.000</td> <td>   -2.531</td> <td>   -2.011</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.39.0]</th>                     <td>   -0.5116</td> <td>    0.128</td> <td>   -3.997</td> <td> 0.000</td> <td>   -0.762</td> <td>   -0.261</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.40.0]</th>                     <td>    0.1774</td> <td>    0.074</td> <td>    2.413</td> <td> 0.016</td> <td>    0.033</td> <td>    0.322</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.41.0]</th>                     <td>   -1.0759</td> <td>    0.170</td> <td>   -6.334</td> <td> 0.000</td> <td>   -1.409</td> <td>   -0.743</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.42.0]</th>                     <td>   -0.4628</td> <td>    0.156</td> <td>   -2.971</td> <td> 0.003</td> <td>   -0.768</td> <td>   -0.157</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.44.0]</th>                     <td>   -0.7887</td> <td>    0.227</td> <td>   -3.482</td> <td> 0.000</td> <td>   -1.233</td> <td>   -0.345</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.45.0]</th>                     <td>   -1.0192</td> <td>    0.097</td> <td>  -10.455</td> <td> 0.000</td> <td>   -1.210</td> <td>   -0.828</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.46.0]</th>                     <td>   -2.5529</td> <td>    0.398</td> <td>   -6.409</td> <td> 0.000</td> <td>   -3.334</td> <td>   -1.772</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.47.0]</th>                     <td>   -0.0144</td> <td>    0.049</td> <td>   -0.292</td> <td> 0.771</td> <td>   -0.111</td> <td>    0.083</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.48.0]</th>                     <td>   -0.7474</td> <td>    0.136</td> <td>   -5.499</td> <td> 0.000</td> <td>   -1.014</td> <td>   -0.481</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.49.0]</th>                     <td>   -1.4253</td> <td>    0.164</td> <td>   -8.715</td> <td> 0.000</td> <td>   -1.746</td> <td>   -1.105</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.50.0]</th>                     <td>   -2.1762</td> <td>    0.228</td> <td>   -9.538</td> <td> 0.000</td> <td>   -2.623</td> <td>   -1.729</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.51.0]</th>                     <td>   -0.8795</td> <td>    0.180</td> <td>   -4.881</td> <td> 0.000</td> <td>   -1.233</td> <td>   -0.526</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.53.0]</th>                     <td>    0.4908</td> <td>    0.055</td> <td>    8.909</td> <td> 0.000</td> <td>    0.383</td> <td>    0.599</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.54.0]</th>                     <td>   -0.3539</td> <td>    0.110</td> <td>   -3.206</td> <td> 0.001</td> <td>   -0.570</td> <td>   -0.138</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.55.0]</th>                     <td>   -0.7085</td> <td>    0.257</td> <td>   -2.756</td> <td> 0.006</td> <td>   -1.212</td> <td>   -0.205</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(fip)[T.56.0]</th>                     <td>   -2.0726</td> <td>    0.197</td> <td>  -10.537</td> <td> 0.000</td> <td>   -2.458</td> <td>   -1.687</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1986.0]</th> <td>    0.0917</td> <td>    0.084</td> <td>    1.092</td> <td> 0.275</td> <td>   -0.073</td> <td>    0.256</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1987.0]</th> <td>   -0.0625</td> <td>    0.152</td> <td>   -0.411</td> <td> 0.681</td> <td>   -0.361</td> <td>    0.236</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1988.0]</th> <td>   -0.2813</td> <td>    0.185</td> <td>   -1.525</td> <td> 0.127</td> <td>   -0.643</td> <td>    0.080</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1989.0]</th> <td>   -0.2215</td> <td>    0.169</td> <td>   -1.310</td> <td> 0.190</td> <td>   -0.553</td> <td>    0.110</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1990.0]</th> <td>   -0.0757</td> <td>    0.258</td> <td>   -0.293</td> <td> 0.769</td> <td>   -0.582</td> <td>    0.431</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1991.0]</th> <td>    0.3830</td> <td>    0.426</td> <td>    0.900</td> <td> 0.368</td> <td>   -0.451</td> <td>    1.217</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1992.0]</th> <td>    0.2526</td> <td>    0.382</td> <td>    0.661</td> <td> 0.509</td> <td>   -0.497</td> <td>    1.002</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1993.0]</th> <td>    0.5158</td> <td>    0.397</td> <td>    1.299</td> <td> 0.194</td> <td>   -0.262</td> <td>    1.294</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1994.0]</th> <td>    0.4472</td> <td>    0.380</td> <td>    1.176</td> <td> 0.240</td> <td>   -0.298</td> <td>    1.193</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1995.0]</th> <td>    0.3197</td> <td>    0.352</td> <td>    0.907</td> <td> 0.364</td> <td>   -0.371</td> <td>    1.010</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1996.0]</th> <td>   -0.3828</td> <td>    0.269</td> <td>   -1.421</td> <td> 0.155</td> <td>   -0.911</td> <td>    0.145</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1997.0]</th> <td>    0.0295</td> <td>    0.282</td> <td>    0.104</td> <td> 0.917</td> <td>   -0.524</td> <td>    0.583</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1998.0]</th> <td>   -0.2078</td> <td>    0.293</td> <td>   -0.710</td> <td> 0.478</td> <td>   -0.782</td> <td>    0.366</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1999.0]</th> <td>    0.0382</td> <td>    0.308</td> <td>    0.124</td> <td> 0.901</td> <td>   -0.566</td> <td>    0.642</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.2000.0]</th> <td>    0.0851</td> <td>    0.262</td> <td>    0.325</td> <td> 0.745</td> <td>   -0.428</td> <td>    0.598</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>acc</th>                                <td>   -0.0004</td> <td>    0.001</td> <td>   -0.357</td> <td> 0.721</td> <td>   -0.002</td> <td>    0.002</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ir</th>                                 <td>    0.0002</td> <td>    0.000</td> <td>    1.274</td> <td> 0.203</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>pi</th>                                 <td> 7.233e-12</td> <td> 3.39e-22</td> <td> 2.13e+10</td> <td> 0.000</td> <td> 7.23e-12</td> <td> 7.23e-12</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>alcohol</th>                            <td>    0.6757</td> <td>    0.171</td> <td>    3.955</td> <td> 0.000</td> <td>    0.341</td> <td>    1.011</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>crack</th>                              <td>   -0.0482</td> <td>    0.033</td> <td>   -1.458</td> <td> 0.145</td> <td>   -0.113</td> <td>    0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>poverty</th>                            <td>   -0.0026</td> <td>    0.010</td> <td>   -0.256</td> <td> 0.798</td> <td>   -0.023</td> <td>    0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>income</th>                             <td>  5.85e-05</td> <td> 2.86e-05</td> <td>    2.044</td> <td> 0.041</td> <td> 2.42e-06</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ur</th>                                 <td>    0.0124</td> <td>    0.026</td> <td>    0.484</td> <td> 0.628</td> <td>   -0.038</td> <td>    0.063</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>61.245</td> <th>  Durbin-Watson:     </th> <td>   1.211</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 319.426</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.050</td> <th>  Prob(JB):          </th> <td>4.34e-70</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 6.232</td> <th>  Cond. No.          </th> <td>9.52e+18</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 3.79e-27. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                    lnr   R-squared:                       0.839\n",
+       "Model:                            OLS   Adj. R-squared:                  0.817\n",
+       "Method:                 Least Squares   F-statistic:                     201.1\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           4.30e-42\n",
+       "Time:                        13:30:28   Log-Likelihood:                -191.07\n",
+       "No. Observations:                 733   AIC:                             558.1\n",
+       "Df Residuals:                     645   BIC:                             962.7\n",
+       "Df Model:                          87                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "======================================================================================================\n",
+       "                                         coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------------------------------\n",
+       "Intercept                              6.6023      0.631     10.456      0.000       5.365       7.840\n",
+       "C(repeal)[T.1.0]                      -1.6162      0.278     -5.818      0.000      -2.161      -1.072\n",
+       "C(year)[T.1986.0]                     -0.0142      0.055     -0.256      0.798      -0.122       0.094\n",
+       "C(year)[T.1987.0]                     -0.1737      0.075     -2.327      0.020      -0.320      -0.027\n",
+       "C(year)[T.1988.0]                     -0.1441      0.095     -1.513      0.130      -0.331       0.043\n",
+       "C(year)[T.1989.0]                     -0.1632      0.134     -1.218      0.223      -0.426       0.099\n",
+       "C(year)[T.1990.0]                     -0.3172      0.150     -2.109      0.035      -0.612      -0.022\n",
+       "C(year)[T.1991.0]                     -0.3716      0.165     -2.246      0.025      -0.696      -0.047\n",
+       "C(year)[T.1992.0]                     -0.5930      0.168     -3.539      0.000      -0.921      -0.265\n",
+       "C(year)[T.1993.0]                     -0.7824      0.195     -4.007      0.000      -1.165      -0.400\n",
+       "C(year)[T.1994.0]                     -0.7746      0.225     -3.439      0.001      -1.216      -0.333\n",
+       "C(year)[T.1995.0]                     -0.9350      0.242     -3.856      0.000      -1.410      -0.460\n",
+       "C(year)[T.1996.0]                     -1.1844      0.280     -4.228      0.000      -1.733      -0.635\n",
+       "C(year)[T.1997.0]                     -1.2330      0.314     -3.925      0.000      -1.849      -0.617\n",
+       "C(year)[T.1998.0]                     -1.1723      0.336     -3.491      0.000      -1.831      -0.514\n",
+       "C(year)[T.1999.0]                     -1.2225      0.389     -3.143      0.002      -1.985      -0.460\n",
+       "C(year)[T.2000.0]                     -1.4156      0.446     -3.177      0.001      -2.289      -0.542\n",
+       "C(fip)[T.2.0]                         -0.8977      0.194     -4.635      0.000      -1.277      -0.518\n",
+       "C(fip)[T.4.0]                         -0.9035      0.168     -5.381      0.000      -1.233      -0.574\n",
+       "C(fip)[T.5.0]                          0.0064      0.045      0.144      0.886      -0.081       0.094\n",
+       "C(fip)[T.6.0]                          0.0840      0.111      0.755      0.450      -0.134       0.302\n",
+       "C(fip)[T.8.0]                         -0.9849      0.241     -4.080      0.000      -1.458      -0.512\n",
+       "C(fip)[T.9.0]                         -1.3984      0.382     -3.664      0.000      -2.146      -0.650\n",
+       "C(fip)[T.10.0]                        -0.9297      0.285     -3.261      0.001      -1.488      -0.371\n",
+       "C(fip)[T.11.0]                        -2.2143      0.583     -3.795      0.000      -3.358      -1.071\n",
+       "C(fip)[T.12.0]                        -0.9462      0.225     -4.211      0.000      -1.387      -0.506\n",
+       "C(fip)[T.13.0]                        -0.4691      0.118     -3.984      0.000      -0.700      -0.238\n",
+       "C(fip)[T.15.0]                        -0.7590      0.097     -7.812      0.000      -0.949      -0.569\n",
+       "C(fip)[T.16.0]                        -1.5459      0.092    -16.826      0.000      -1.726      -1.366\n",
+       "C(fip)[T.17.0]                        -0.7829      0.233     -3.363      0.001      -1.239      -0.327\n",
+       "C(fip)[T.18.0]                        -0.2189      0.114     -1.920      0.055      -0.442       0.005\n",
+       "C(fip)[T.19.0]                        -0.3851      0.204     -1.887      0.059      -0.785       0.015\n",
+       "C(fip)[T.20.0]                         0.1033      0.119      0.870      0.384      -0.129       0.336\n",
+       "C(fip)[T.21.0]                        -0.1655      0.049     -3.396      0.001      -0.261      -0.070\n",
+       "C(fip)[T.22.0]                        -0.8190      0.107     -7.626      0.000      -1.030      -0.609\n",
+       "C(fip)[T.23.0]                        -2.2270      0.140    -15.852      0.000      -2.502      -1.952\n",
+       "C(fip)[T.24.0]                        -1.3288      0.260     -5.110      0.000      -1.839      -0.819\n",
+       "C(fip)[T.25.0]                        -1.5452      0.312     -4.951      0.000      -2.157      -0.933\n",
+       "C(fip)[T.26.0]                        -0.9182      0.178     -5.158      0.000      -1.267      -0.569\n",
+       "C(fip)[T.27.0]                        -0.7059      0.235     -3.002      0.003      -1.167      -0.245\n",
+       "C(fip)[T.28.0]                        -0.0630      0.084     -0.752      0.452      -0.227       0.101\n",
+       "C(fip)[T.29.0]                        -0.1710      0.123     -1.388      0.165      -0.413       0.071\n",
+       "C(fip)[T.30.0]                        -1.5919      0.158    -10.102      0.000      -1.901      -1.283\n",
+       "C(fip)[T.31.0]                        -0.2417      0.149     -1.623      0.105      -0.534       0.050\n",
+       "C(fip)[T.32.0]                        -2.4190      0.499     -4.848      0.000      -3.397      -1.441\n",
+       "C(fip)[T.33.0]                        -3.9837      0.533     -7.474      0.000      -5.028      -2.939\n",
+       "C(fip)[T.34.0]                        -2.1379      0.334     -6.408      0.000      -2.792      -1.484\n",
+       "C(fip)[T.35.0]                        -1.1200      0.107    -10.453      0.000      -1.330      -0.910\n",
+       "C(fip)[T.36.0]                        -0.5342      0.104     -5.149      0.000      -0.738      -0.331\n",
+       "C(fip)[T.37.0]                        -0.0919      0.084     -1.092      0.275      -0.257       0.073\n",
+       "C(fip)[T.38.0]                        -2.2710      0.133    -17.115      0.000      -2.531      -2.011\n",
+       "C(fip)[T.39.0]                        -0.5116      0.128     -3.997      0.000      -0.762      -0.261\n",
+       "C(fip)[T.40.0]                         0.1774      0.074      2.413      0.016       0.033       0.322\n",
+       "C(fip)[T.41.0]                        -1.0759      0.170     -6.334      0.000      -1.409      -0.743\n",
+       "C(fip)[T.42.0]                        -0.4628      0.156     -2.971      0.003      -0.768      -0.157\n",
+       "C(fip)[T.44.0]                        -0.7887      0.227     -3.482      0.000      -1.233      -0.345\n",
+       "C(fip)[T.45.0]                        -1.0192      0.097    -10.455      0.000      -1.210      -0.828\n",
+       "C(fip)[T.46.0]                        -2.5529      0.398     -6.409      0.000      -3.334      -1.772\n",
+       "C(fip)[T.47.0]                        -0.0144      0.049     -0.292      0.771      -0.111       0.083\n",
+       "C(fip)[T.48.0]                        -0.7474      0.136     -5.499      0.000      -1.014      -0.481\n",
+       "C(fip)[T.49.0]                        -1.4253      0.164     -8.715      0.000      -1.746      -1.105\n",
+       "C(fip)[T.50.0]                        -2.1762      0.228     -9.538      0.000      -2.623      -1.729\n",
+       "C(fip)[T.51.0]                        -0.8795      0.180     -4.881      0.000      -1.233      -0.526\n",
+       "C(fip)[T.53.0]                         0.4908      0.055      8.909      0.000       0.383       0.599\n",
+       "C(fip)[T.54.0]                        -0.3539      0.110     -3.206      0.001      -0.570      -0.138\n",
+       "C(fip)[T.55.0]                        -0.7085      0.257     -2.756      0.006      -1.212      -0.205\n",
+       "C(fip)[T.56.0]                        -2.0726      0.197    -10.537      0.000      -2.458      -1.687\n",
+       "C(repeal)[T.1.0]:C(year)[T.1986.0]     0.0917      0.084      1.092      0.275      -0.073       0.256\n",
+       "C(repeal)[T.1.0]:C(year)[T.1987.0]    -0.0625      0.152     -0.411      0.681      -0.361       0.236\n",
+       "C(repeal)[T.1.0]:C(year)[T.1988.0]    -0.2813      0.185     -1.525      0.127      -0.643       0.080\n",
+       "C(repeal)[T.1.0]:C(year)[T.1989.0]    -0.2215      0.169     -1.310      0.190      -0.553       0.110\n",
+       "C(repeal)[T.1.0]:C(year)[T.1990.0]    -0.0757      0.258     -0.293      0.769      -0.582       0.431\n",
+       "C(repeal)[T.1.0]:C(year)[T.1991.0]     0.3830      0.426      0.900      0.368      -0.451       1.217\n",
+       "C(repeal)[T.1.0]:C(year)[T.1992.0]     0.2526      0.382      0.661      0.509      -0.497       1.002\n",
+       "C(repeal)[T.1.0]:C(year)[T.1993.0]     0.5158      0.397      1.299      0.194      -0.262       1.294\n",
+       "C(repeal)[T.1.0]:C(year)[T.1994.0]     0.4472      0.380      1.176      0.240      -0.298       1.193\n",
+       "C(repeal)[T.1.0]:C(year)[T.1995.0]     0.3197      0.352      0.907      0.364      -0.371       1.010\n",
+       "C(repeal)[T.1.0]:C(year)[T.1996.0]    -0.3828      0.269     -1.421      0.155      -0.911       0.145\n",
+       "C(repeal)[T.1.0]:C(year)[T.1997.0]     0.0295      0.282      0.104      0.917      -0.524       0.583\n",
+       "C(repeal)[T.1.0]:C(year)[T.1998.0]    -0.2078      0.293     -0.710      0.478      -0.782       0.366\n",
+       "C(repeal)[T.1.0]:C(year)[T.1999.0]     0.0382      0.308      0.124      0.901      -0.566       0.642\n",
+       "C(repeal)[T.1.0]:C(year)[T.2000.0]     0.0851      0.262      0.325      0.745      -0.428       0.598\n",
+       "acc                                   -0.0004      0.001     -0.357      0.721      -0.002       0.002\n",
+       "ir                                     0.0002      0.000      1.274      0.203      -0.000       0.000\n",
+       "pi                                  7.233e-12   3.39e-22   2.13e+10      0.000    7.23e-12    7.23e-12\n",
+       "alcohol                                0.6757      0.171      3.955      0.000       0.341       1.011\n",
+       "crack                                 -0.0482      0.033     -1.458      0.145      -0.113       0.017\n",
+       "poverty                               -0.0026      0.010     -0.256      0.798      -0.023       0.017\n",
+       "income                               5.85e-05   2.86e-05      2.044      0.041    2.42e-06       0.000\n",
+       "ur                                     0.0124      0.026      0.484      0.628      -0.038       0.063\n",
+       "==============================================================================\n",
+       "Omnibus:                       61.245   Durbin-Watson:                   1.211\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              319.426\n",
+       "Skew:                          -0.050   Prob(JB):                     4.34e-70\n",
+       "Kurtosis:                       6.232   Cond. No.                     9.52e+18\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 3.79e-27. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "abortion_filt = abortion[(abortion.race == 2) & (abortion.sex == 2) & (abortion.age == 20)]\n",
+    "\n",
+    "reg = sm.OLS.from_formula(\"\"\"lnr ~ C(repeal)*C(year) + C(fip) + acc + ir + pi + alcohol+ crack + poverty+ income+ ur\"\"\",\n",
+    "            data = abortion_filt, freq_weights = abortion_filt['totpop']).fit(\n",
+    "            cov_type='cluster', cov_kwds={'groups': abortion_filt['fip']})\n",
+    "reg.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 39, but rank is 27\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>lnr</td>       <th>  R-squared:         </th> <td>   0.304</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.285</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   60.98</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>1.21e-29</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:30:28</td>     <th>  Log-Likelihood:    </th> <td> -1512.5</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  1435</td>      <th>  AIC:               </th> <td>   3103.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>  1396</td>      <th>  BIC:               </th> <td>   3309.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    38</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "                   <td></td>                     <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>                          <td>    7.6205</td> <td>    0.964</td> <td>    7.908</td> <td> 0.000</td> <td>    5.732</td> <td>    9.509</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]</th>                   <td>   -0.5738</td> <td>    0.395</td> <td>   -1.454</td> <td> 0.146</td> <td>   -1.347</td> <td>    0.200</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1986.0]</th>                  <td>   -0.0373</td> <td>    0.084</td> <td>   -0.446</td> <td> 0.656</td> <td>   -0.202</td> <td>    0.127</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1987.0]</th>                  <td>   -0.1494</td> <td>    0.103</td> <td>   -1.452</td> <td> 0.146</td> <td>   -0.351</td> <td>    0.052</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1988.0]</th>                  <td>   -0.1116</td> <td>    0.121</td> <td>   -0.920</td> <td> 0.357</td> <td>   -0.349</td> <td>    0.126</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1989.0]</th>                  <td>   -0.1295</td> <td>    0.142</td> <td>   -0.912</td> <td> 0.362</td> <td>   -0.408</td> <td>    0.149</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1990.0]</th>                  <td>   -0.2513</td> <td>    0.176</td> <td>   -1.426</td> <td> 0.154</td> <td>   -0.597</td> <td>    0.094</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1991.0]</th>                  <td>   -0.3977</td> <td>    0.157</td> <td>   -2.538</td> <td> 0.011</td> <td>   -0.705</td> <td>   -0.091</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1992.0]</th>                  <td>   -0.5652</td> <td>    0.174</td> <td>   -3.250</td> <td> 0.001</td> <td>   -0.906</td> <td>   -0.224</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1993.0]</th>                  <td>   -0.8746</td> <td>    0.214</td> <td>   -4.080</td> <td> 0.000</td> <td>   -1.295</td> <td>   -0.454</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1994.0]</th>                  <td>   -0.8592</td> <td>    0.229</td> <td>   -3.756</td> <td> 0.000</td> <td>   -1.307</td> <td>   -0.411</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1995.0]</th>                  <td>   -1.0074</td> <td>    0.233</td> <td>   -4.318</td> <td> 0.000</td> <td>   -1.465</td> <td>   -0.550</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1996.0]</th>                  <td>   -1.1561</td> <td>    0.259</td> <td>   -4.459</td> <td> 0.000</td> <td>   -1.664</td> <td>   -0.648</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1997.0]</th>                  <td>   -1.2460</td> <td>    0.294</td> <td>   -4.231</td> <td> 0.000</td> <td>   -1.823</td> <td>   -0.669</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1998.0]</th>                  <td>   -1.0992</td> <td>    0.325</td> <td>   -3.382</td> <td> 0.001</td> <td>   -1.736</td> <td>   -0.462</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.1999.0]</th>                  <td>   -1.1717</td> <td>    0.347</td> <td>   -3.381</td> <td> 0.001</td> <td>   -1.851</td> <td>   -0.492</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2000.0]</th>                  <td>   -1.3523</td> <td>    0.425</td> <td>   -3.183</td> <td> 0.001</td> <td>   -2.185</td> <td>   -0.520</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1986.0]</th> <td>    0.1810</td> <td>    0.109</td> <td>    1.659</td> <td> 0.097</td> <td>   -0.033</td> <td>    0.395</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1987.0]</th> <td>   -0.0042</td> <td>    0.166</td> <td>   -0.026</td> <td> 0.980</td> <td>   -0.329</td> <td>    0.321</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1988.0]</th> <td>   -0.2271</td> <td>    0.216</td> <td>   -1.053</td> <td> 0.292</td> <td>   -0.650</td> <td>    0.195</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1989.0]</th> <td>   -0.2291</td> <td>    0.178</td> <td>   -1.285</td> <td> 0.199</td> <td>   -0.579</td> <td>    0.120</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1990.0]</th> <td>   -0.3368</td> <td>    0.159</td> <td>   -2.119</td> <td> 0.034</td> <td>   -0.648</td> <td>   -0.025</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1991.0]</th> <td>    0.0718</td> <td>    0.360</td> <td>    0.199</td> <td> 0.842</td> <td>   -0.635</td> <td>    0.778</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1992.0]</th> <td>   -0.1311</td> <td>    0.336</td> <td>   -0.390</td> <td> 0.696</td> <td>   -0.789</td> <td>    0.527</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1993.0]</th> <td>    0.2365</td> <td>    0.423</td> <td>    0.559</td> <td> 0.576</td> <td>   -0.593</td> <td>    1.066</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1994.0]</th> <td>    0.1473</td> <td>    0.409</td> <td>    0.360</td> <td> 0.719</td> <td>   -0.654</td> <td>    0.949</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1995.0]</th> <td>    0.0381</td> <td>    0.369</td> <td>    0.103</td> <td> 0.918</td> <td>   -0.685</td> <td>    0.761</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1996.0]</th> <td>   -0.5589</td> <td>    0.368</td> <td>   -1.518</td> <td> 0.129</td> <td>   -1.281</td> <td>    0.163</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1997.0]</th> <td>   -0.3572</td> <td>    0.288</td> <td>   -1.242</td> <td> 0.214</td> <td>   -0.921</td> <td>    0.206</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1998.0]</th> <td>   -0.7028</td> <td>    0.406</td> <td>   -1.730</td> <td> 0.084</td> <td>   -1.499</td> <td>    0.093</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.1999.0]</th> <td>   -0.2976</td> <td>    0.362</td> <td>   -0.822</td> <td> 0.411</td> <td>   -1.007</td> <td>    0.412</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(repeal)[T.1.0]:C(year)[T.2000.0]</th> <td>   -0.2860</td> <td>    0.267</td> <td>   -1.072</td> <td> 0.284</td> <td>   -0.809</td> <td>    0.237</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>acc</th>                                <td>   -0.0012</td> <td>    0.002</td> <td>   -0.671</td> <td> 0.502</td> <td>   -0.005</td> <td>    0.002</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ir</th>                                 <td>    0.0001</td> <td>    0.000</td> <td>    0.922</td> <td> 0.356</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>pi</th>                                 <td>-5.854e-17</td> <td> 3.55e-17</td> <td>   -1.647</td> <td> 0.099</td> <td>-1.28e-16</td> <td> 1.11e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>alcohol</th>                            <td>   -0.1050</td> <td>    0.157</td> <td>   -0.668</td> <td> 0.504</td> <td>   -0.413</td> <td>    0.203</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>crack</th>                              <td>   -0.0852</td> <td>    0.049</td> <td>   -1.740</td> <td> 0.082</td> <td>   -0.181</td> <td>    0.011</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>poverty</th>                            <td>    0.0260</td> <td>    0.025</td> <td>    1.052</td> <td> 0.293</td> <td>   -0.022</td> <td>    0.074</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>income</th>                             <td> 3.308e-05</td> <td> 3.39e-05</td> <td>    0.975</td> <td> 0.330</td> <td>-3.34e-05</td> <td> 9.96e-05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>ur</th>                                 <td>   -0.0195</td> <td>    0.043</td> <td>   -0.455</td> <td> 0.649</td> <td>   -0.104</td> <td>    0.065</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>71.173</td> <th>  Durbin-Watson:     </th> <td>   1.311</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  81.091</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.553</td> <th>  Prob(JB):          </th> <td>2.46e-18</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 3.366</td> <th>  Cond. No.          </th> <td>1.03e+16</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 6.3e-21. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                    lnr   R-squared:                       0.304\n",
+       "Model:                            OLS   Adj. R-squared:                  0.285\n",
+       "Method:                 Least Squares   F-statistic:                     60.98\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           1.21e-29\n",
+       "Time:                        13:30:28   Log-Likelihood:                -1512.5\n",
+       "No. Observations:                1435   AIC:                             3103.\n",
+       "Df Residuals:                    1396   BIC:                             3309.\n",
+       "Df Model:                          38                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "======================================================================================================\n",
+       "                                         coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------------------------------\n",
+       "Intercept                              7.6205      0.964      7.908      0.000       5.732       9.509\n",
+       "C(repeal)[T.1.0]                      -0.5738      0.395     -1.454      0.146      -1.347       0.200\n",
+       "C(year)[T.1986.0]                     -0.0373      0.084     -0.446      0.656      -0.202       0.127\n",
+       "C(year)[T.1987.0]                     -0.1494      0.103     -1.452      0.146      -0.351       0.052\n",
+       "C(year)[T.1988.0]                     -0.1116      0.121     -0.920      0.357      -0.349       0.126\n",
+       "C(year)[T.1989.0]                     -0.1295      0.142     -0.912      0.362      -0.408       0.149\n",
+       "C(year)[T.1990.0]                     -0.2513      0.176     -1.426      0.154      -0.597       0.094\n",
+       "C(year)[T.1991.0]                     -0.3977      0.157     -2.538      0.011      -0.705      -0.091\n",
+       "C(year)[T.1992.0]                     -0.5652      0.174     -3.250      0.001      -0.906      -0.224\n",
+       "C(year)[T.1993.0]                     -0.8746      0.214     -4.080      0.000      -1.295      -0.454\n",
+       "C(year)[T.1994.0]                     -0.8592      0.229     -3.756      0.000      -1.307      -0.411\n",
+       "C(year)[T.1995.0]                     -1.0074      0.233     -4.318      0.000      -1.465      -0.550\n",
+       "C(year)[T.1996.0]                     -1.1561      0.259     -4.459      0.000      -1.664      -0.648\n",
+       "C(year)[T.1997.0]                     -1.2460      0.294     -4.231      0.000      -1.823      -0.669\n",
+       "C(year)[T.1998.0]                     -1.0992      0.325     -3.382      0.001      -1.736      -0.462\n",
+       "C(year)[T.1999.0]                     -1.1717      0.347     -3.381      0.001      -1.851      -0.492\n",
+       "C(year)[T.2000.0]                     -1.3523      0.425     -3.183      0.001      -2.185      -0.520\n",
+       "C(repeal)[T.1.0]:C(year)[T.1986.0]     0.1810      0.109      1.659      0.097      -0.033       0.395\n",
+       "C(repeal)[T.1.0]:C(year)[T.1987.0]    -0.0042      0.166     -0.026      0.980      -0.329       0.321\n",
+       "C(repeal)[T.1.0]:C(year)[T.1988.0]    -0.2271      0.216     -1.053      0.292      -0.650       0.195\n",
+       "C(repeal)[T.1.0]:C(year)[T.1989.0]    -0.2291      0.178     -1.285      0.199      -0.579       0.120\n",
+       "C(repeal)[T.1.0]:C(year)[T.1990.0]    -0.3368      0.159     -2.119      0.034      -0.648      -0.025\n",
+       "C(repeal)[T.1.0]:C(year)[T.1991.0]     0.0718      0.360      0.199      0.842      -0.635       0.778\n",
+       "C(repeal)[T.1.0]:C(year)[T.1992.0]    -0.1311      0.336     -0.390      0.696      -0.789       0.527\n",
+       "C(repeal)[T.1.0]:C(year)[T.1993.0]     0.2365      0.423      0.559      0.576      -0.593       1.066\n",
+       "C(repeal)[T.1.0]:C(year)[T.1994.0]     0.1473      0.409      0.360      0.719      -0.654       0.949\n",
+       "C(repeal)[T.1.0]:C(year)[T.1995.0]     0.0381      0.369      0.103      0.918      -0.685       0.761\n",
+       "C(repeal)[T.1.0]:C(year)[T.1996.0]    -0.5589      0.368     -1.518      0.129      -1.281       0.163\n",
+       "C(repeal)[T.1.0]:C(year)[T.1997.0]    -0.3572      0.288     -1.242      0.214      -0.921       0.206\n",
+       "C(repeal)[T.1.0]:C(year)[T.1998.0]    -0.7028      0.406     -1.730      0.084      -1.499       0.093\n",
+       "C(repeal)[T.1.0]:C(year)[T.1999.0]    -0.2976      0.362     -0.822      0.411      -1.007       0.412\n",
+       "C(repeal)[T.1.0]:C(year)[T.2000.0]    -0.2860      0.267     -1.072      0.284      -0.809       0.237\n",
+       "acc                                   -0.0012      0.002     -0.671      0.502      -0.005       0.002\n",
+       "ir                                     0.0001      0.000      0.922      0.356      -0.000       0.000\n",
+       "pi                                 -5.854e-17   3.55e-17     -1.647      0.099   -1.28e-16    1.11e-17\n",
+       "alcohol                               -0.1050      0.157     -0.668      0.504      -0.413       0.203\n",
+       "crack                                 -0.0852      0.049     -1.740      0.082      -0.181       0.011\n",
+       "poverty                                0.0260      0.025      1.052      0.293      -0.022       0.074\n",
+       "income                              3.308e-05   3.39e-05      0.975      0.330   -3.34e-05    9.96e-05\n",
+       "ur                                    -0.0195      0.043     -0.455      0.649      -0.104       0.065\n",
+       "==============================================================================\n",
+       "Omnibus:                       71.173   Durbin-Watson:                   1.311\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               81.091\n",
+       "Skew:                          -0.553   Prob(JB):                     2.46e-18\n",
+       "Kurtosis:                       3.366   Cond. No.                     1.03e+16\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 6.3e-21. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "abortion_filt = abortion[(abortion.bf == 1) & abortion.age.isin([20,25])]\n",
+    "\n",
+    "regddd = sm.OLS.from_formula(\"\"\"lnr ~ C(repeal)*C(year) + acc + ir + pi + alcohol + crack + poverty + income + ur\"\"\",\n",
+    "            data = abortion_filt, freq_weights = abortion_filt['totpop']).fit(\n",
+    "            cov_type='cluster', cov_kwds={'groups': abortion_filt['fip']})\n",
+    "regddd.summary()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Why did we suggest that conducting this additional analysis not conducted in the original study?\n",
+    "- How convinced are you now of the abortion selection hypothesis?  Why/why not?\n",
+    "- Could you have concluded this had you not exploited all of the testable predictions of the original table showing roll out across cohort and time?  \n",
+    "- How important was our ``model`` to forming testable predictions and falsifications? \n",
+    "\n",
+    "## Cheng and Hoekstra (2013)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "castle = read_data('castle.dta')\n",
+    "crime1 = (\"jhcitizen_c\", \"jhpolice_c\", \n",
+    "            \"murder\", \"homicide\", \n",
+    "            \"robbery\", \"assault\", \"burglary\",\n",
+    "            \"larceny\", \"motor\", \"robbery_gun_r\")\n",
+    "demo = (\"emo\", \"blackm_15_24\", \"whitem_15_24\", \n",
+    "          \"blackm_25_44\", \"whitem_25_44\")\n",
+    "\n",
+    "# variables dropped to prevent colinearity\n",
+    "dropped_vars = (\"r20004\", \"r20014\",\n",
+    "                  \"r20024\", \"r20034\",\n",
+    "                  \"r20044\", \"r20054\",\n",
+    "                  \"r20064\", \"r20074\",\n",
+    "                  \"r20084\", \"r20094\",\n",
+    "                  \"r20101\", \"r20102\", \"r20103\",\n",
+    "                  \"r20104\", \"trend_9\", \"trend_46\",\n",
+    "                  \"trend_49\", \"trend_50\", \"trend_51\")\n",
+    "cols = pd.Series(castle.columns)\n",
+    "trend_cols = set(cols[cols.str.contains('^trend')])\n",
+    "lintrend = castle[trend_cols - set(dropped_vars)]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 167, but rank is 48\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>       <td>l_homicide</td>    <th>  R-squared:         </th> <td>   0.953</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.936</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   1313.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>1.44e-63</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:30:29</td>     <th>  Log-Likelihood:    </th> <td>  351.46</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   550</td>      <th>  AIC:               </th> <td>  -404.9</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   401</td>      <th>  BIC:               </th> <td>   237.3</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>   148</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "          <td></td>            <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>        <td>    7.2406</td> <td>    3.995</td> <td>    1.812</td> <td> 0.070</td> <td>   -0.589</td> <td>   15.071</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2001]</th>  <td>    0.8388</td> <td>    0.465</td> <td>    1.803</td> <td> 0.071</td> <td>   -0.073</td> <td>    1.751</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2002]</th>  <td>    0.6755</td> <td>    0.386</td> <td>    1.752</td> <td> 0.080</td> <td>   -0.080</td> <td>    1.431</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2003]</th>  <td>    0.5801</td> <td>    0.316</td> <td>    1.835</td> <td> 0.067</td> <td>   -0.040</td> <td>    1.200</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2004]</th>  <td>    0.4553</td> <td>    0.244</td> <td>    1.865</td> <td> 0.062</td> <td>   -0.023</td> <td>    0.934</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2005]</th>  <td>    0.3475</td> <td>    0.172</td> <td>    2.015</td> <td> 0.044</td> <td>    0.009</td> <td>    0.685</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2006]</th>  <td>    0.2448</td> <td>    0.106</td> <td>    2.312</td> <td> 0.021</td> <td>    0.037</td> <td>    0.452</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2007]</th>  <td>    0.0931</td> <td>    0.031</td> <td>    2.987</td> <td> 0.003</td> <td>    0.032</td> <td>    0.154</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2008]</th>  <td>   -0.0802</td> <td>    0.058</td> <td>   -1.378</td> <td> 0.168</td> <td>   -0.194</td> <td>    0.034</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2009]</th>  <td>   -0.2938</td> <td>    0.149</td> <td>   -1.975</td> <td> 0.048</td> <td>   -0.585</td> <td>   -0.002</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2010]</th>  <td>   -0.4554</td> <td>    0.238</td> <td>   -1.910</td> <td> 0.056</td> <td>   -0.923</td> <td>    0.012</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.2]</th>      <td>    1.2878</td> <td>    0.433</td> <td>    2.975</td> <td> 0.003</td> <td>    0.439</td> <td>    2.136</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.3]</th>      <td>    1.5196</td> <td>    0.352</td> <td>    4.313</td> <td> 0.000</td> <td>    0.829</td> <td>    2.210</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.4]</th>      <td>   -0.1818</td> <td>    0.189</td> <td>   -0.964</td> <td> 0.335</td> <td>   -0.551</td> <td>    0.188</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.5]</th>      <td>    1.3348</td> <td>    0.323</td> <td>    4.130</td> <td> 0.000</td> <td>    0.701</td> <td>    1.968</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.6]</th>      <td>    0.5522</td> <td>    0.337</td> <td>    1.636</td> <td> 0.102</td> <td>   -0.109</td> <td>    1.214</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.7]</th>      <td>    1.4516</td> <td>    0.755</td> <td>    1.923</td> <td> 0.055</td> <td>   -0.028</td> <td>    2.931</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.8]</th>      <td>   -0.9708</td> <td>    0.243</td> <td>   -4.000</td> <td> 0.000</td> <td>   -1.446</td> <td>   -0.495</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.10]</th>     <td>   -0.4353</td> <td>    0.177</td> <td>   -2.460</td> <td> 0.014</td> <td>   -0.782</td> <td>   -0.089</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.11]</th>     <td>    0.4953</td> <td>    0.310</td> <td>    1.596</td> <td> 0.111</td> <td>   -0.113</td> <td>    1.104</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.12]</th>     <td>    0.3626</td> <td>    0.398</td> <td>    0.910</td> <td> 0.363</td> <td>   -0.418</td> <td>    1.144</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.13]</th>     <td>   -2.6831</td> <td>    1.543</td> <td>   -1.739</td> <td> 0.082</td> <td>   -5.707</td> <td>    0.341</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.14]</th>     <td>    1.6046</td> <td>    0.399</td> <td>    4.017</td> <td> 0.000</td> <td>    0.822</td> <td>    2.387</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.15]</th>     <td>    1.2927</td> <td>    0.434</td> <td>    2.978</td> <td> 0.003</td> <td>    0.442</td> <td>    2.143</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.16]</th>     <td>   -0.2032</td> <td>    0.385</td> <td>   -0.528</td> <td> 0.597</td> <td>   -0.957</td> <td>    0.551</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.17]</th>     <td>    0.7915</td> <td>    0.284</td> <td>    2.785</td> <td> 0.005</td> <td>    0.235</td> <td>    1.348</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.18]</th>     <td>   -1.0180</td> <td>    0.206</td> <td>   -4.947</td> <td> 0.000</td> <td>   -1.421</td> <td>   -0.615</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.19]</th>     <td>    1.1794</td> <td>    0.289</td> <td>    4.086</td> <td> 0.000</td> <td>    0.614</td> <td>    1.745</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.20]</th>     <td>   -2.4746</td> <td>    0.995</td> <td>   -2.487</td> <td> 0.013</td> <td>   -4.425</td> <td>   -0.525</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.21]</th>     <td>    0.5090</td> <td>    0.308</td> <td>    1.653</td> <td> 0.098</td> <td>   -0.095</td> <td>    1.113</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.22]</th>     <td>    0.4280</td> <td>    0.542</td> <td>    0.790</td> <td> 0.429</td> <td>   -0.634</td> <td>    1.490</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.23]</th>     <td>    1.6261</td> <td>    0.457</td> <td>    3.556</td> <td> 0.000</td> <td>    0.730</td> <td>    2.522</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.24]</th>     <td>   -0.1563</td> <td>    0.455</td> <td>   -0.344</td> <td> 0.731</td> <td>   -1.048</td> <td>    0.735</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.25]</th>     <td>    0.4456</td> <td>    0.223</td> <td>    2.002</td> <td> 0.045</td> <td>    0.009</td> <td>    0.882</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.26]</th>     <td>    1.5548</td> <td>    0.488</td> <td>    3.189</td> <td> 0.001</td> <td>    0.599</td> <td>    2.510</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.27]</th>     <td>    0.0203</td> <td>    0.320</td> <td>    0.063</td> <td> 0.949</td> <td>   -0.608</td> <td>    0.648</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.28]</th>     <td>    0.4219</td> <td>    0.349</td> <td>    1.207</td> <td> 0.227</td> <td>   -0.263</td> <td>    1.107</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.29]</th>     <td>    1.5162</td> <td>    0.414</td> <td>    3.659</td> <td> 0.000</td> <td>    0.704</td> <td>    2.328</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.30]</th>     <td>   -0.8745</td> <td>    0.543</td> <td>   -1.612</td> <td> 0.107</td> <td>   -1.938</td> <td>    0.189</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.31]</th>     <td>    1.6273</td> <td>    0.808</td> <td>    2.015</td> <td> 0.044</td> <td>    0.045</td> <td>    3.210</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.32]</th>     <td>   -1.6308</td> <td>    1.915</td> <td>   -0.852</td> <td> 0.394</td> <td>   -5.384</td> <td>    2.123</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.33]</th>     <td>    1.8652</td> <td>    0.665</td> <td>    2.803</td> <td> 0.005</td> <td>    0.561</td> <td>    3.169</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.34]</th>     <td>    0.1139</td> <td>    0.377</td> <td>    0.302</td> <td> 0.763</td> <td>   -0.626</td> <td>    0.854</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.35]</th>     <td>   -4.6254</td> <td>    2.701</td> <td>   -1.712</td> <td> 0.087</td> <td>   -9.919</td> <td>    0.669</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.36]</th>     <td>    0.9980</td> <td>    0.445</td> <td>    2.245</td> <td> 0.025</td> <td>    0.127</td> <td>    1.869</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.37]</th>     <td>    0.0716</td> <td>    0.313</td> <td>    0.229</td> <td> 0.819</td> <td>   -0.542</td> <td>    0.685</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.38]</th>     <td>   -0.5893</td> <td>    0.154</td> <td>   -3.818</td> <td> 0.000</td> <td>   -0.892</td> <td>   -0.287</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.39]</th>     <td>    1.9239</td> <td>    0.835</td> <td>    2.304</td> <td> 0.021</td> <td>    0.287</td> <td>    3.560</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.40]</th>     <td>   -0.7604</td> <td>    0.614</td> <td>   -1.238</td> <td> 0.216</td> <td>   -1.964</td> <td>    0.443</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.41]</th>     <td>    0.4543</td> <td>    0.317</td> <td>    1.434</td> <td> 0.151</td> <td>   -0.167</td> <td>    1.075</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.42]</th>     <td>   -2.9779</td> <td>    1.336</td> <td>   -2.230</td> <td> 0.026</td> <td>   -5.596</td> <td>   -0.360</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.43]</th>     <td>   -0.7387</td> <td>    0.345</td> <td>   -2.143</td> <td> 0.032</td> <td>   -1.414</td> <td>   -0.063</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.44]</th>     <td>    0.4509</td> <td>    0.318</td> <td>    1.420</td> <td> 0.156</td> <td>   -0.172</td> <td>    1.073</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.45]</th>     <td>    0.0519</td> <td>    0.447</td> <td>    0.116</td> <td> 0.908</td> <td>   -0.825</td> <td>    0.929</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.46]</th>     <td>   -4.0563</td> <td>    2.597</td> <td>   -1.562</td> <td> 0.118</td> <td>   -9.147</td> <td>    1.034</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.47]</th>     <td>    0.0528</td> <td>    0.341</td> <td>    0.155</td> <td> 0.877</td> <td>   -0.615</td> <td>    0.721</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.48]</th>     <td>    0.2189</td> <td>    0.412</td> <td>    0.532</td> <td> 0.595</td> <td>   -0.588</td> <td>    1.026</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.49]</th>     <td>   -1.0181</td> <td>    0.526</td> <td>   -1.935</td> <td> 0.053</td> <td>   -2.049</td> <td>    0.013</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.50]</th>     <td>    0.7409</td> <td>    0.397</td> <td>    1.864</td> <td> 0.062</td> <td>   -0.038</td> <td>    1.520</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.51]</th>     <td>    0.0829</td> <td>    0.382</td> <td>    0.217</td> <td> 0.828</td> <td>   -0.666</td> <td>    0.832</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>blackm_15_24</th>     <td>    0.0967</td> <td>    0.140</td> <td>    0.692</td> <td> 0.489</td> <td>   -0.177</td> <td>    0.371</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>whitem_15_24</th>     <td>    0.0659</td> <td>    0.044</td> <td>    1.512</td> <td> 0.131</td> <td>   -0.020</td> <td>    0.151</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>blackm_25_44</th>     <td>    0.1188</td> <td>    0.147</td> <td>    0.810</td> <td> 0.418</td> <td>   -0.169</td> <td>    0.406</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>whitem_25_44</th>     <td>   -0.0370</td> <td>    0.014</td> <td>   -2.575</td> <td> 0.010</td> <td>   -0.065</td> <td>   -0.009</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_exp_subsidy</th>    <td>   -0.0698</td> <td>    0.064</td> <td>   -1.089</td> <td> 0.276</td> <td>   -0.195</td> <td>    0.056</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_exp_pubwelfare</th> <td>    0.0593</td> <td>    0.112</td> <td>    0.527</td> <td> 0.598</td> <td>   -0.161</td> <td>    0.280</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_police</th>         <td>    0.1354</td> <td>    0.076</td> <td>    1.784</td> <td> 0.074</td> <td>   -0.013</td> <td>    0.284</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>unemployrt</th>       <td>    0.0188</td> <td>    0.020</td> <td>    0.925</td> <td> 0.355</td> <td>   -0.021</td> <td>    0.059</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>poverty</th>          <td>   -0.0513</td> <td>    0.024</td> <td>   -2.164</td> <td> 0.030</td> <td>   -0.098</td> <td>   -0.005</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_income</th>         <td>   -0.5838</td> <td>    0.307</td> <td>   -1.900</td> <td> 0.057</td> <td>   -1.186</td> <td>    0.018</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_prisoner</th>       <td>   -0.1321</td> <td>    0.366</td> <td>   -0.361</td> <td> 0.718</td> <td>   -0.850</td> <td>    0.585</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>l_lagprisoner</th>    <td>   -0.2081</td> <td>    0.424</td> <td>   -0.490</td> <td> 0.624</td> <td>   -1.040</td> <td>    0.624</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20023</th>           <td>    0.9233</td> <td>    0.224</td> <td>    4.119</td> <td> 0.000</td> <td>    0.484</td> <td>    1.363</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20072</th>           <td>    0.0580</td> <td>    0.048</td> <td>    1.208</td> <td> 0.227</td> <td>   -0.036</td> <td>    0.152</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20002</th>           <td>    1.0233</td> <td>    0.600</td> <td>    1.706</td> <td> 0.088</td> <td>   -0.153</td> <td>    2.199</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20081</th>           <td>    0.1750</td> <td>    0.050</td> <td>    3.526</td> <td> 0.000</td> <td>    0.078</td> <td>    0.272</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20001</th>           <td>    0.3567</td> <td>    0.805</td> <td>    0.443</td> <td> 0.658</td> <td>   -1.221</td> <td>    1.934</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20103</th>           <td>   -0.6226</td> <td>    0.137</td> <td>   -4.561</td> <td> 0.000</td> <td>   -0.890</td> <td>   -0.355</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20024</th>           <td>    0.2431</td> <td>    0.077</td> <td>    3.161</td> <td> 0.002</td> <td>    0.092</td> <td>    0.394</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20034</th>           <td>    0.2085</td> <td>    0.090</td> <td>    2.325</td> <td> 0.020</td> <td>    0.033</td> <td>    0.384</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20073</th>           <td>    0.0517</td> <td>    0.033</td> <td>    1.550</td> <td> 0.121</td> <td>   -0.014</td> <td>    0.117</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20071</th>           <td>   -0.1047</td> <td>    0.044</td> <td>   -2.366</td> <td> 0.018</td> <td>   -0.192</td> <td>   -0.018</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20031</th>           <td>   -0.3509</td> <td>    0.214</td> <td>   -1.643</td> <td> 0.100</td> <td>   -0.769</td> <td>    0.068</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20094</th>           <td>   -0.1061</td> <td>    0.051</td> <td>   -2.079</td> <td> 0.038</td> <td>   -0.206</td> <td>   -0.006</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20054</th>           <td>    0.1116</td> <td>    0.055</td> <td>    2.039</td> <td> 0.041</td> <td>    0.004</td> <td>    0.219</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20013</th>           <td>    1.0684</td> <td>    0.265</td> <td>    4.035</td> <td> 0.000</td> <td>    0.549</td> <td>    1.587</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20061</th>           <td>   -0.0583</td> <td>    0.049</td> <td>   -1.178</td> <td> 0.239</td> <td>   -0.155</td> <td>    0.039</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20063</th>           <td>    0.1880</td> <td>    0.054</td> <td>    3.511</td> <td> 0.000</td> <td>    0.083</td> <td>    0.293</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20051</th>           <td>   -0.1812</td> <td>    0.066</td> <td>   -2.747</td> <td> 0.006</td> <td>   -0.310</td> <td>   -0.052</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20101</th>           <td>    0.3865</td> <td>    0.172</td> <td>    2.248</td> <td> 0.025</td> <td>    0.050</td> <td>    0.723</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20082</th>           <td>    0.0248</td> <td>    0.066</td> <td>    0.377</td> <td> 0.706</td> <td>   -0.104</td> <td>    0.153</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20033</th>           <td>    0.7257</td> <td>    0.182</td> <td>    3.994</td> <td> 0.000</td> <td>    0.370</td> <td>    1.082</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20014</th>           <td>    0.3141</td> <td>    0.120</td> <td>    2.628</td> <td> 0.009</td> <td>    0.080</td> <td>    0.548</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20022</th>           <td>   -0.0461</td> <td>    0.100</td> <td>   -0.462</td> <td> 0.644</td> <td>   -0.242</td> <td>    0.149</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20052</th>           <td>    0.0224</td> <td>    0.048</td> <td>    0.468</td> <td> 0.640</td> <td>   -0.072</td> <td>    0.116</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20044</th>           <td>    0.2064</td> <td>    0.069</td> <td>    3.000</td> <td> 0.003</td> <td>    0.072</td> <td>    0.341</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20092</th>           <td>    0.0142</td> <td>    0.066</td> <td>    0.214</td> <td> 0.831</td> <td>   -0.116</td> <td>    0.144</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20004</th>           <td>    1.2081</td> <td>    0.649</td> <td>    1.862</td> <td> 0.063</td> <td>   -0.063</td> <td>    2.480</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20053</th>           <td>    0.3946</td> <td>    0.096</td> <td>    4.118</td> <td> 0.000</td> <td>    0.207</td> <td>    0.582</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20043</th>           <td>    0.5333</td> <td>    0.140</td> <td>    3.807</td> <td> 0.000</td> <td>    0.259</td> <td>    0.808</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20021</th>           <td>   -0.4448</td> <td>    0.267</td> <td>   -1.663</td> <td> 0.096</td> <td>   -0.969</td> <td>    0.079</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20003</th>           <td>    2.2469</td> <td>    0.779</td> <td>    2.884</td> <td> 0.004</td> <td>    0.720</td> <td>    3.774</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20104</th>           <td>   -0.1947</td> <td>    0.081</td> <td>   -2.403</td> <td> 0.016</td> <td>   -0.354</td> <td>   -0.036</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20032</th>           <td>   -0.0032</td> <td>    0.064</td> <td>   -0.050</td> <td> 0.960</td> <td>   -0.129</td> <td>    0.123</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20074</th>           <td>    0.0881</td> <td>    0.056</td> <td>    1.577</td> <td> 0.115</td> <td>   -0.021</td> <td>    0.198</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20102</th>           <td>   -0.0245</td> <td>    0.072</td> <td>   -0.340</td> <td> 0.734</td> <td>   -0.166</td> <td>    0.117</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20011</th>           <td>   -0.5378</td> <td>    0.269</td> <td>   -1.998</td> <td> 0.046</td> <td>   -1.065</td> <td>   -0.010</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20041</th>           <td>   -0.2474</td> <td>    0.164</td> <td>   -1.508</td> <td> 0.131</td> <td>   -0.569</td> <td>    0.074</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20091</th>           <td>    0.1370</td> <td>    0.127</td> <td>    1.082</td> <td> 0.279</td> <td>   -0.111</td> <td>    0.385</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20012</th>           <td>   -0.0058</td> <td>    0.074</td> <td>   -0.079</td> <td> 0.937</td> <td>   -0.152</td> <td>    0.140</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20093</th>           <td>   -0.3390</td> <td>    0.090</td> <td>   -3.751</td> <td> 0.000</td> <td>   -0.516</td> <td>   -0.162</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20062</th>           <td>    0.0415</td> <td>    0.051</td> <td>    0.822</td> <td> 0.411</td> <td>   -0.057</td> <td>    0.141</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20084</th>           <td>   -0.1085</td> <td>    0.038</td> <td>   -2.883</td> <td> 0.004</td> <td>   -0.182</td> <td>   -0.035</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20083</th>           <td>   -0.1715</td> <td>    0.055</td> <td>   -3.144</td> <td> 0.002</td> <td>   -0.278</td> <td>   -0.065</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20064</th>           <td>    0.0736</td> <td>    0.045</td> <td>    1.625</td> <td> 0.104</td> <td>   -0.015</td> <td>    0.162</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20042</th>           <td>   -0.0370</td> <td>    0.060</td> <td>   -0.617</td> <td> 0.537</td> <td>   -0.154</td> <td>    0.080</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_37</th>         <td>    0.3359</td> <td>    0.113</td> <td>    2.970</td> <td> 0.003</td> <td>    0.114</td> <td>    0.558</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_40</th>         <td>   -0.0286</td> <td>    0.107</td> <td>   -0.268</td> <td> 0.788</td> <td>   -0.237</td> <td>    0.180</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_26</th>         <td>    0.1428</td> <td>    0.080</td> <td>    1.791</td> <td> 0.073</td> <td>   -0.013</td> <td>    0.299</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_43</th>         <td>    0.2971</td> <td>    0.106</td> <td>    2.795</td> <td> 0.005</td> <td>    0.089</td> <td>    0.505</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_22</th>         <td>    0.0627</td> <td>    0.106</td> <td>    0.592</td> <td> 0.554</td> <td>   -0.145</td> <td>    0.270</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_2</th>          <td>    0.1369</td> <td>    0.088</td> <td>    1.547</td> <td> 0.122</td> <td>   -0.037</td> <td>    0.310</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_29</th>         <td>    0.1509</td> <td>    0.086</td> <td>    1.753</td> <td> 0.080</td> <td>   -0.018</td> <td>    0.320</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_30</th>         <td>   -0.0537</td> <td>    0.102</td> <td>   -0.526</td> <td> 0.599</td> <td>   -0.253</td> <td>    0.146</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_50</th>         <td>    0.1100</td> <td>    0.085</td> <td>    1.300</td> <td> 0.194</td> <td>   -0.056</td> <td>    0.276</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_21</th>         <td>    0.2855</td> <td>    0.103</td> <td>    2.770</td> <td> 0.006</td> <td>    0.083</td> <td>    0.487</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_46</th>         <td>   -0.0413</td> <td>    0.090</td> <td>   -0.459</td> <td> 0.646</td> <td>   -0.218</td> <td>    0.135</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_3</th>          <td>    0.1701</td> <td>    0.093</td> <td>    1.824</td> <td> 0.068</td> <td>   -0.013</td> <td>    0.353</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_14</th>         <td>    0.0932</td> <td>    0.081</td> <td>    1.145</td> <td> 0.252</td> <td>   -0.066</td> <td>    0.253</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_20</th>         <td>    0.0756</td> <td>    0.100</td> <td>    0.758</td> <td> 0.448</td> <td>   -0.120</td> <td>    0.271</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_16</th>         <td>    0.1367</td> <td>    0.087</td> <td>    1.576</td> <td> 0.115</td> <td>   -0.033</td> <td>    0.307</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_33</th>         <td>    0.0038</td> <td>    0.104</td> <td>    0.037</td> <td> 0.971</td> <td>   -0.200</td> <td>    0.208</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_44</th>         <td>    0.2974</td> <td>    0.105</td> <td>    2.839</td> <td> 0.005</td> <td>    0.092</td> <td>    0.503</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_5</th>          <td>    0.1450</td> <td>    0.084</td> <td>    1.734</td> <td> 0.083</td> <td>   -0.019</td> <td>    0.309</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_9</th>          <td>-1.053e-15</td> <td> 3.91e-28</td> <td> -2.7e+12</td> <td> 0.000</td> <td>-1.05e-15</td> <td>-1.05e-15</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_31</th>         <td>    0.0449</td> <td>    0.106</td> <td>    0.425</td> <td> 0.671</td> <td>   -0.162</td> <td>    0.252</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_18</th>         <td>    0.3409</td> <td>    0.116</td> <td>    2.931</td> <td> 0.003</td> <td>    0.113</td> <td>    0.569</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_51</th>         <td>    0.1666</td> <td>    0.093</td> <td>    1.793</td> <td> 0.073</td> <td>   -0.016</td> <td>    0.349</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_48</th>         <td>    0.1834</td> <td>    0.096</td> <td>    1.916</td> <td> 0.055</td> <td>   -0.004</td> <td>    0.371</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_8</th>          <td>    0.3719</td> <td>    0.101</td> <td>    3.680</td> <td> 0.000</td> <td>    0.174</td> <td>    0.570</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_7</th>          <td>    0.0453</td> <td>    0.114</td> <td>    0.398</td> <td> 0.691</td> <td>   -0.178</td> <td>    0.269</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_13</th>         <td>    0.1862</td> <td>    0.091</td> <td>    2.057</td> <td> 0.040</td> <td>    0.009</td> <td>    0.364</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_28</th>         <td>    0.1358</td> <td>    0.089</td> <td>    1.534</td> <td> 0.125</td> <td>   -0.038</td> <td>    0.309</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_35</th>         <td>    0.1415</td> <td>    0.083</td> <td>    1.695</td> <td> 0.090</td> <td>   -0.022</td> <td>    0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_42</th>         <td>    0.2757</td> <td>    0.079</td> <td>    3.485</td> <td> 0.000</td> <td>    0.121</td> <td>    0.431</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_1</th>          <td>    0.2971</td> <td>    0.112</td> <td>    2.658</td> <td> 0.008</td> <td>    0.078</td> <td>    0.516</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_25</th>         <td>    0.2703</td> <td>    0.108</td> <td>    2.497</td> <td> 0.013</td> <td>    0.058</td> <td>    0.482</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_49</th>         <td>    0.3914</td> <td>    0.130</td> <td>    3.019</td> <td> 0.003</td> <td>    0.137</td> <td>    0.646</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_10</th>         <td>    0.3259</td> <td>    0.113</td> <td>    2.891</td> <td> 0.004</td> <td>    0.105</td> <td>    0.547</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_27</th>         <td>    0.1912</td> <td>    0.092</td> <td>    2.073</td> <td> 0.038</td> <td>    0.010</td> <td>    0.372</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_19</th>         <td>    0.3105</td> <td>    0.111</td> <td>    2.792</td> <td> 0.005</td> <td>    0.093</td> <td>    0.528</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_41</th>         <td>    0.3036</td> <td>    0.103</td> <td>    2.941</td> <td> 0.003</td> <td>    0.101</td> <td>    0.506</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_38</th>         <td>    0.1776</td> <td>    0.096</td> <td>    1.854</td> <td> 0.064</td> <td>   -0.010</td> <td>    0.365</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_24</th>         <td>    0.1026</td> <td>    0.092</td> <td>    1.121</td> <td> 0.262</td> <td>   -0.077</td> <td>    0.282</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_32</th>         <td>    0.1751</td> <td>    0.088</td> <td>    1.997</td> <td> 0.046</td> <td>    0.003</td> <td>    0.347</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_34</th>         <td>    0.2983</td> <td>    0.115</td> <td>    2.602</td> <td> 0.009</td> <td>    0.074</td> <td>    0.523</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_11</th>         <td>    0.2891</td> <td>    0.107</td> <td>    2.710</td> <td> 0.007</td> <td>    0.080</td> <td>    0.498</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_17</th>         <td>    0.1150</td> <td>    0.079</td> <td>    1.457</td> <td> 0.145</td> <td>   -0.040</td> <td>    0.270</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_45</th>         <td>    0.1557</td> <td>    0.091</td> <td>    1.711</td> <td> 0.087</td> <td>   -0.023</td> <td>    0.334</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_36</th>         <td>    0.1567</td> <td>    0.082</td> <td>    1.910</td> <td> 0.056</td> <td>   -0.004</td> <td>    0.317</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_23</th>         <td>    0.1206</td> <td>    0.077</td> <td>    1.568</td> <td> 0.117</td> <td>   -0.030</td> <td>    0.271</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_12</th>         <td>    0.1299</td> <td>    0.091</td> <td>    1.436</td> <td> 0.151</td> <td>   -0.047</td> <td>    0.307</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_6</th>          <td>    0.1723</td> <td>    0.098</td> <td>    1.765</td> <td> 0.078</td> <td>   -0.019</td> <td>    0.364</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_15</th>         <td>    0.1297</td> <td>    0.088</td> <td>    1.479</td> <td> 0.139</td> <td>   -0.042</td> <td>    0.301</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_4</th>          <td>    0.3230</td> <td>    0.116</td> <td>    2.786</td> <td> 0.005</td> <td>    0.096</td> <td>    0.550</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_39</th>         <td>    0.0609</td> <td>    0.118</td> <td>    0.517</td> <td> 0.605</td> <td>   -0.170</td> <td>    0.292</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>trend_47</th>         <td>    0.3008</td> <td>    0.112</td> <td>    2.679</td> <td> 0.007</td> <td>    0.081</td> <td>    0.521</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>post</th>             <td>    0.0176</td> <td>    0.052</td> <td>    0.336</td> <td> 0.737</td> <td>   -0.085</td> <td>    0.120</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>45.652</td> <th>  Durbin-Watson:     </th> <td>   2.468</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 124.426</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.395</td> <th>  Prob(JB):          </th> <td>9.58e-28</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 5.192</td> <th>  Cond. No.          </th> <td>6.91e+17</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 2.84e-30. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:             l_homicide   R-squared:                       0.953\n",
+       "Model:                            OLS   Adj. R-squared:                  0.936\n",
+       "Method:                 Least Squares   F-statistic:                     1313.\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           1.44e-63\n",
+       "Time:                        13:30:29   Log-Likelihood:                 351.46\n",
+       "No. Observations:                 550   AIC:                            -404.9\n",
+       "Df Residuals:                     401   BIC:                             237.3\n",
+       "Df Model:                         148                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "====================================================================================\n",
+       "                       coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------------\n",
+       "Intercept            7.2406      3.995      1.812      0.070      -0.589      15.071\n",
+       "C(year)[T.2001]      0.8388      0.465      1.803      0.071      -0.073       1.751\n",
+       "C(year)[T.2002]      0.6755      0.386      1.752      0.080      -0.080       1.431\n",
+       "C(year)[T.2003]      0.5801      0.316      1.835      0.067      -0.040       1.200\n",
+       "C(year)[T.2004]      0.4553      0.244      1.865      0.062      -0.023       0.934\n",
+       "C(year)[T.2005]      0.3475      0.172      2.015      0.044       0.009       0.685\n",
+       "C(year)[T.2006]      0.2448      0.106      2.312      0.021       0.037       0.452\n",
+       "C(year)[T.2007]      0.0931      0.031      2.987      0.003       0.032       0.154\n",
+       "C(year)[T.2008]     -0.0802      0.058     -1.378      0.168      -0.194       0.034\n",
+       "C(year)[T.2009]     -0.2938      0.149     -1.975      0.048      -0.585      -0.002\n",
+       "C(year)[T.2010]     -0.4554      0.238     -1.910      0.056      -0.923       0.012\n",
+       "C(sid)[T.2]          1.2878      0.433      2.975      0.003       0.439       2.136\n",
+       "C(sid)[T.3]          1.5196      0.352      4.313      0.000       0.829       2.210\n",
+       "C(sid)[T.4]         -0.1818      0.189     -0.964      0.335      -0.551       0.188\n",
+       "C(sid)[T.5]          1.3348      0.323      4.130      0.000       0.701       1.968\n",
+       "C(sid)[T.6]          0.5522      0.337      1.636      0.102      -0.109       1.214\n",
+       "C(sid)[T.7]          1.4516      0.755      1.923      0.055      -0.028       2.931\n",
+       "C(sid)[T.8]         -0.9708      0.243     -4.000      0.000      -1.446      -0.495\n",
+       "C(sid)[T.10]        -0.4353      0.177     -2.460      0.014      -0.782      -0.089\n",
+       "C(sid)[T.11]         0.4953      0.310      1.596      0.111      -0.113       1.104\n",
+       "C(sid)[T.12]         0.3626      0.398      0.910      0.363      -0.418       1.144\n",
+       "C(sid)[T.13]        -2.6831      1.543     -1.739      0.082      -5.707       0.341\n",
+       "C(sid)[T.14]         1.6046      0.399      4.017      0.000       0.822       2.387\n",
+       "C(sid)[T.15]         1.2927      0.434      2.978      0.003       0.442       2.143\n",
+       "C(sid)[T.16]        -0.2032      0.385     -0.528      0.597      -0.957       0.551\n",
+       "C(sid)[T.17]         0.7915      0.284      2.785      0.005       0.235       1.348\n",
+       "C(sid)[T.18]        -1.0180      0.206     -4.947      0.000      -1.421      -0.615\n",
+       "C(sid)[T.19]         1.1794      0.289      4.086      0.000       0.614       1.745\n",
+       "C(sid)[T.20]        -2.4746      0.995     -2.487      0.013      -4.425      -0.525\n",
+       "C(sid)[T.21]         0.5090      0.308      1.653      0.098      -0.095       1.113\n",
+       "C(sid)[T.22]         0.4280      0.542      0.790      0.429      -0.634       1.490\n",
+       "C(sid)[T.23]         1.6261      0.457      3.556      0.000       0.730       2.522\n",
+       "C(sid)[T.24]        -0.1563      0.455     -0.344      0.731      -1.048       0.735\n",
+       "C(sid)[T.25]         0.4456      0.223      2.002      0.045       0.009       0.882\n",
+       "C(sid)[T.26]         1.5548      0.488      3.189      0.001       0.599       2.510\n",
+       "C(sid)[T.27]         0.0203      0.320      0.063      0.949      -0.608       0.648\n",
+       "C(sid)[T.28]         0.4219      0.349      1.207      0.227      -0.263       1.107\n",
+       "C(sid)[T.29]         1.5162      0.414      3.659      0.000       0.704       2.328\n",
+       "C(sid)[T.30]        -0.8745      0.543     -1.612      0.107      -1.938       0.189\n",
+       "C(sid)[T.31]         1.6273      0.808      2.015      0.044       0.045       3.210\n",
+       "C(sid)[T.32]        -1.6308      1.915     -0.852      0.394      -5.384       2.123\n",
+       "C(sid)[T.33]         1.8652      0.665      2.803      0.005       0.561       3.169\n",
+       "C(sid)[T.34]         0.1139      0.377      0.302      0.763      -0.626       0.854\n",
+       "C(sid)[T.35]        -4.6254      2.701     -1.712      0.087      -9.919       0.669\n",
+       "C(sid)[T.36]         0.9980      0.445      2.245      0.025       0.127       1.869\n",
+       "C(sid)[T.37]         0.0716      0.313      0.229      0.819      -0.542       0.685\n",
+       "C(sid)[T.38]        -0.5893      0.154     -3.818      0.000      -0.892      -0.287\n",
+       "C(sid)[T.39]         1.9239      0.835      2.304      0.021       0.287       3.560\n",
+       "C(sid)[T.40]        -0.7604      0.614     -1.238      0.216      -1.964       0.443\n",
+       "C(sid)[T.41]         0.4543      0.317      1.434      0.151      -0.167       1.075\n",
+       "C(sid)[T.42]        -2.9779      1.336     -2.230      0.026      -5.596      -0.360\n",
+       "C(sid)[T.43]        -0.7387      0.345     -2.143      0.032      -1.414      -0.063\n",
+       "C(sid)[T.44]         0.4509      0.318      1.420      0.156      -0.172       1.073\n",
+       "C(sid)[T.45]         0.0519      0.447      0.116      0.908      -0.825       0.929\n",
+       "C(sid)[T.46]        -4.0563      2.597     -1.562      0.118      -9.147       1.034\n",
+       "C(sid)[T.47]         0.0528      0.341      0.155      0.877      -0.615       0.721\n",
+       "C(sid)[T.48]         0.2189      0.412      0.532      0.595      -0.588       1.026\n",
+       "C(sid)[T.49]        -1.0181      0.526     -1.935      0.053      -2.049       0.013\n",
+       "C(sid)[T.50]         0.7409      0.397      1.864      0.062      -0.038       1.520\n",
+       "C(sid)[T.51]         0.0829      0.382      0.217      0.828      -0.666       0.832\n",
+       "blackm_15_24         0.0967      0.140      0.692      0.489      -0.177       0.371\n",
+       "whitem_15_24         0.0659      0.044      1.512      0.131      -0.020       0.151\n",
+       "blackm_25_44         0.1188      0.147      0.810      0.418      -0.169       0.406\n",
+       "whitem_25_44        -0.0370      0.014     -2.575      0.010      -0.065      -0.009\n",
+       "l_exp_subsidy       -0.0698      0.064     -1.089      0.276      -0.195       0.056\n",
+       "l_exp_pubwelfare     0.0593      0.112      0.527      0.598      -0.161       0.280\n",
+       "l_police             0.1354      0.076      1.784      0.074      -0.013       0.284\n",
+       "unemployrt           0.0188      0.020      0.925      0.355      -0.021       0.059\n",
+       "poverty             -0.0513      0.024     -2.164      0.030      -0.098      -0.005\n",
+       "l_income            -0.5838      0.307     -1.900      0.057      -1.186       0.018\n",
+       "l_prisoner          -0.1321      0.366     -0.361      0.718      -0.850       0.585\n",
+       "l_lagprisoner       -0.2081      0.424     -0.490      0.624      -1.040       0.624\n",
+       "r20023               0.9233      0.224      4.119      0.000       0.484       1.363\n",
+       "r20072               0.0580      0.048      1.208      0.227      -0.036       0.152\n",
+       "r20002               1.0233      0.600      1.706      0.088      -0.153       2.199\n",
+       "r20081               0.1750      0.050      3.526      0.000       0.078       0.272\n",
+       "r20001               0.3567      0.805      0.443      0.658      -1.221       1.934\n",
+       "r20103              -0.6226      0.137     -4.561      0.000      -0.890      -0.355\n",
+       "r20024               0.2431      0.077      3.161      0.002       0.092       0.394\n",
+       "r20034               0.2085      0.090      2.325      0.020       0.033       0.384\n",
+       "r20073               0.0517      0.033      1.550      0.121      -0.014       0.117\n",
+       "r20071              -0.1047      0.044     -2.366      0.018      -0.192      -0.018\n",
+       "r20031              -0.3509      0.214     -1.643      0.100      -0.769       0.068\n",
+       "r20094              -0.1061      0.051     -2.079      0.038      -0.206      -0.006\n",
+       "r20054               0.1116      0.055      2.039      0.041       0.004       0.219\n",
+       "r20013               1.0684      0.265      4.035      0.000       0.549       1.587\n",
+       "r20061              -0.0583      0.049     -1.178      0.239      -0.155       0.039\n",
+       "r20063               0.1880      0.054      3.511      0.000       0.083       0.293\n",
+       "r20051              -0.1812      0.066     -2.747      0.006      -0.310      -0.052\n",
+       "r20101               0.3865      0.172      2.248      0.025       0.050       0.723\n",
+       "r20082               0.0248      0.066      0.377      0.706      -0.104       0.153\n",
+       "r20033               0.7257      0.182      3.994      0.000       0.370       1.082\n",
+       "r20014               0.3141      0.120      2.628      0.009       0.080       0.548\n",
+       "r20022              -0.0461      0.100     -0.462      0.644      -0.242       0.149\n",
+       "r20052               0.0224      0.048      0.468      0.640      -0.072       0.116\n",
+       "r20044               0.2064      0.069      3.000      0.003       0.072       0.341\n",
+       "r20092               0.0142      0.066      0.214      0.831      -0.116       0.144\n",
+       "r20004               1.2081      0.649      1.862      0.063      -0.063       2.480\n",
+       "r20053               0.3946      0.096      4.118      0.000       0.207       0.582\n",
+       "r20043               0.5333      0.140      3.807      0.000       0.259       0.808\n",
+       "r20021              -0.4448      0.267     -1.663      0.096      -0.969       0.079\n",
+       "r20003               2.2469      0.779      2.884      0.004       0.720       3.774\n",
+       "r20104              -0.1947      0.081     -2.403      0.016      -0.354      -0.036\n",
+       "r20032              -0.0032      0.064     -0.050      0.960      -0.129       0.123\n",
+       "r20074               0.0881      0.056      1.577      0.115      -0.021       0.198\n",
+       "r20102              -0.0245      0.072     -0.340      0.734      -0.166       0.117\n",
+       "r20011              -0.5378      0.269     -1.998      0.046      -1.065      -0.010\n",
+       "r20041              -0.2474      0.164     -1.508      0.131      -0.569       0.074\n",
+       "r20091               0.1370      0.127      1.082      0.279      -0.111       0.385\n",
+       "r20012              -0.0058      0.074     -0.079      0.937      -0.152       0.140\n",
+       "r20093              -0.3390      0.090     -3.751      0.000      -0.516      -0.162\n",
+       "r20062               0.0415      0.051      0.822      0.411      -0.057       0.141\n",
+       "r20084              -0.1085      0.038     -2.883      0.004      -0.182      -0.035\n",
+       "r20083              -0.1715      0.055     -3.144      0.002      -0.278      -0.065\n",
+       "r20064               0.0736      0.045      1.625      0.104      -0.015       0.162\n",
+       "r20042              -0.0370      0.060     -0.617      0.537      -0.154       0.080\n",
+       "trend_37             0.3359      0.113      2.970      0.003       0.114       0.558\n",
+       "trend_40            -0.0286      0.107     -0.268      0.788      -0.237       0.180\n",
+       "trend_26             0.1428      0.080      1.791      0.073      -0.013       0.299\n",
+       "trend_43             0.2971      0.106      2.795      0.005       0.089       0.505\n",
+       "trend_22             0.0627      0.106      0.592      0.554      -0.145       0.270\n",
+       "trend_2              0.1369      0.088      1.547      0.122      -0.037       0.310\n",
+       "trend_29             0.1509      0.086      1.753      0.080      -0.018       0.320\n",
+       "trend_30            -0.0537      0.102     -0.526      0.599      -0.253       0.146\n",
+       "trend_50             0.1100      0.085      1.300      0.194      -0.056       0.276\n",
+       "trend_21             0.2855      0.103      2.770      0.006       0.083       0.487\n",
+       "trend_46            -0.0413      0.090     -0.459      0.646      -0.218       0.135\n",
+       "trend_3              0.1701      0.093      1.824      0.068      -0.013       0.353\n",
+       "trend_14             0.0932      0.081      1.145      0.252      -0.066       0.253\n",
+       "trend_20             0.0756      0.100      0.758      0.448      -0.120       0.271\n",
+       "trend_16             0.1367      0.087      1.576      0.115      -0.033       0.307\n",
+       "trend_33             0.0038      0.104      0.037      0.971      -0.200       0.208\n",
+       "trend_44             0.2974      0.105      2.839      0.005       0.092       0.503\n",
+       "trend_5              0.1450      0.084      1.734      0.083      -0.019       0.309\n",
+       "trend_9          -1.053e-15   3.91e-28   -2.7e+12      0.000   -1.05e-15   -1.05e-15\n",
+       "trend_31             0.0449      0.106      0.425      0.671      -0.162       0.252\n",
+       "trend_18             0.3409      0.116      2.931      0.003       0.113       0.569\n",
+       "trend_51             0.1666      0.093      1.793      0.073      -0.016       0.349\n",
+       "trend_48             0.1834      0.096      1.916      0.055      -0.004       0.371\n",
+       "trend_8              0.3719      0.101      3.680      0.000       0.174       0.570\n",
+       "trend_7              0.0453      0.114      0.398      0.691      -0.178       0.269\n",
+       "trend_13             0.1862      0.091      2.057      0.040       0.009       0.364\n",
+       "trend_28             0.1358      0.089      1.534      0.125      -0.038       0.309\n",
+       "trend_35             0.1415      0.083      1.695      0.090      -0.022       0.305\n",
+       "trend_42             0.2757      0.079      3.485      0.000       0.121       0.431\n",
+       "trend_1              0.2971      0.112      2.658      0.008       0.078       0.516\n",
+       "trend_25             0.2703      0.108      2.497      0.013       0.058       0.482\n",
+       "trend_49             0.3914      0.130      3.019      0.003       0.137       0.646\n",
+       "trend_10             0.3259      0.113      2.891      0.004       0.105       0.547\n",
+       "trend_27             0.1912      0.092      2.073      0.038       0.010       0.372\n",
+       "trend_19             0.3105      0.111      2.792      0.005       0.093       0.528\n",
+       "trend_41             0.3036      0.103      2.941      0.003       0.101       0.506\n",
+       "trend_38             0.1776      0.096      1.854      0.064      -0.010       0.365\n",
+       "trend_24             0.1026      0.092      1.121      0.262      -0.077       0.282\n",
+       "trend_32             0.1751      0.088      1.997      0.046       0.003       0.347\n",
+       "trend_34             0.2983      0.115      2.602      0.009       0.074       0.523\n",
+       "trend_11             0.2891      0.107      2.710      0.007       0.080       0.498\n",
+       "trend_17             0.1150      0.079      1.457      0.145      -0.040       0.270\n",
+       "trend_45             0.1557      0.091      1.711      0.087      -0.023       0.334\n",
+       "trend_36             0.1567      0.082      1.910      0.056      -0.004       0.317\n",
+       "trend_23             0.1206      0.077      1.568      0.117      -0.030       0.271\n",
+       "trend_12             0.1299      0.091      1.436      0.151      -0.047       0.307\n",
+       "trend_6              0.1723      0.098      1.765      0.078      -0.019       0.364\n",
+       "trend_15             0.1297      0.088      1.479      0.139      -0.042       0.301\n",
+       "trend_4              0.3230      0.116      2.786      0.005       0.096       0.550\n",
+       "trend_39             0.0609      0.118      0.517      0.605      -0.170       0.292\n",
+       "trend_47             0.3008      0.112      2.679      0.007       0.081       0.521\n",
+       "post                 0.0176      0.052      0.336      0.737      -0.085       0.120\n",
+       "==============================================================================\n",
+       "Omnibus:                       45.652   Durbin-Watson:                   2.468\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              124.426\n",
+       "Skew:                          -0.395   Prob(JB):                     9.58e-28\n",
+       "Kurtosis:                       5.192   Cond. No.                     6.91e+17\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 2.84e-30. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "region = set(cols[cols.str.contains('^r20')])\n",
+    "lintrend = set(cols[cols.str.contains('^trend')])\n",
+    "\n",
+    "exocrime = (\"l_lacerny\", \"l_motor\")\n",
+    "spending = (\"l_exp_subsidy\", \"l_exp_pubwelfare\")\n",
+    "xvar = (\n",
+    "  \"blackm_15_24\", \"whitem_15_24\", \"blackm_25_44\", \"whitem_25_44\",\n",
+    "  \"l_exp_subsidy\", \"l_exp_pubwelfare\",\n",
+    "  \"l_police\", \"unemployrt\", \"poverty\", \n",
+    "  \"l_income\", \"l_prisoner\", \"l_lagprisoner\"\n",
+    ")\n",
+    "\n",
+    "law = (\"cdl\")\n",
+    "\n",
+    "dd_formula = \"l_homicide ~ {} + {} + {} + post + C(year) + C(sid)\".format(\n",
+    "    \"+\".join(xvar), \n",
+    "    \"+\".join(region),\n",
+    "    \"+\".join(lintrend))\n",
+    "\n",
+    "#Fixed effect regression using post as treatment variable \n",
+    "dd_reg = sm.OLS.from_formula(dd_formula,\n",
+    "            data = castle, freq_weights = castle['popwt']).fit(cov_type='cluster', cov_kwds={'groups':castle['sid']})\n",
+    "dd_reg.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTIONS\n",
+    "\n",
+    "- What effect does this analysis say reforming castle doctrine laws has on homicides?\n",
+    "- What are the key parts of these legislative reforms that you think may be causing this result?\n",
+    "- Explain what SUTVA requires in order for these estimates to be causal?\n",
+    "- Assume there are spillovers to neighboring states created by castle doctrine reforms.  Does that imply that Cheng and Hoekstra's result is too large or too small?  Why/why not?\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 88, but rank is 20\n",
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/regression/linear_model.py:1817: RuntimeWarning: invalid value encountered in sqrt\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>       <td>l_homicide</td>    <th>  R-squared:         </th> <td>   0.900</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.862</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>1.369e+13</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>3.99e-127</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:30:29</td>     <th>  Log-Likelihood:    </th> <td>  88.377</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>   231</td>      <th>  AIC:               </th> <td>  -48.75</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   167</td>      <th>  BIC:               </th> <td>   171.6</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    63</td>      <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "           <td></td>              <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>           <td>    1.4335</td> <td>    0.028</td> <td>   50.373</td> <td> 0.000</td> <td>    1.378</td> <td>    1.489</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-8.0]</th> <td>    0.0251</td> <td>    0.225</td> <td>    0.112</td> <td> 0.911</td> <td>   -0.417</td> <td>    0.467</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-7.0]</th> <td>   -0.0308</td> <td>    0.172</td> <td>   -0.179</td> <td> 0.858</td> <td>   -0.368</td> <td>    0.306</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-6.0]</th> <td>    0.2006</td> <td>    0.076</td> <td>    2.630</td> <td> 0.009</td> <td>    0.051</td> <td>    0.350</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-5.0]</th> <td>    0.1677</td> <td>    0.040</td> <td>    4.187</td> <td> 0.000</td> <td>    0.089</td> <td>    0.246</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-4.0]</th> <td>    0.1493</td> <td>    0.072</td> <td>    2.064</td> <td> 0.039</td> <td>    0.008</td> <td>    0.291</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-3.0]</th> <td>    0.1816</td> <td>    0.051</td> <td>    3.578</td> <td> 0.000</td> <td>    0.082</td> <td>    0.281</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-2.0]</th> <td>    0.1777</td> <td>    0.059</td> <td>    3.001</td> <td> 0.003</td> <td>    0.062</td> <td>    0.294</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.-1.0]</th> <td>    0.0910</td> <td>    0.059</td> <td>    1.532</td> <td> 0.125</td> <td>   -0.025</td> <td>    0.207</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.0.0]</th>  <td>    0.2041</td> <td>    0.081</td> <td>    2.509</td> <td> 0.012</td> <td>    0.045</td> <td>    0.364</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.1.0]</th>  <td>    0.1575</td> <td>    0.060</td> <td>    2.645</td> <td> 0.008</td> <td>    0.041</td> <td>    0.274</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.2.0]</th>  <td>    0.2202</td> <td>    0.065</td> <td>    3.391</td> <td> 0.001</td> <td>    0.093</td> <td>    0.347</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.3.0]</th>  <td>    0.1171</td> <td>    0.065</td> <td>    1.812</td> <td> 0.070</td> <td>   -0.010</td> <td>    0.244</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.4.0]</th>  <td>    0.0682</td> <td>    0.083</td> <td>    0.823</td> <td> 0.410</td> <td>   -0.094</td> <td>    0.230</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(time_til)[T.5.0]</th>  <td>    0.1440</td> <td>    0.053</td> <td>    2.738</td> <td> 0.006</td> <td>    0.041</td> <td>    0.247</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2001]</th>     <td>    0.1401</td> <td>    0.058</td> <td>    2.413</td> <td> 0.016</td> <td>    0.026</td> <td>    0.254</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2002]</th>     <td>    0.0441</td> <td>    0.037</td> <td>    1.179</td> <td> 0.238</td> <td>   -0.029</td> <td>    0.118</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2003]</th>     <td>    0.1269</td> <td>    0.026</td> <td>    4.832</td> <td> 0.000</td> <td>    0.075</td> <td>    0.178</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2004]</th>     <td>    0.1348</td> <td>    0.042</td> <td>    3.191</td> <td> 0.001</td> <td>    0.052</td> <td>    0.218</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2005]</th>     <td>    0.1113</td> <td>    0.063</td> <td>    1.754</td> <td> 0.079</td> <td>   -0.013</td> <td>    0.236</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2006]</th>     <td>    0.1687</td> <td>    0.058</td> <td>    2.927</td> <td> 0.003</td> <td>    0.056</td> <td>    0.282</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2007]</th>     <td>    0.1696</td> <td>    0.069</td> <td>    2.460</td> <td> 0.014</td> <td>    0.035</td> <td>    0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2008]</th>     <td>    0.1001</td> <td>    0.055</td> <td>    1.810</td> <td> 0.070</td> <td>   -0.008</td> <td>    0.209</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2009]</th>     <td>    0.1079</td> <td>    0.052</td> <td>    2.083</td> <td> 0.037</td> <td>    0.006</td> <td>    0.209</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(year)[T.2010]</th>     <td>    0.0697</td> <td>    0.044</td> <td>    1.596</td> <td> 0.110</td> <td>   -0.016</td> <td>    0.155</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.2]</th>         <td>   -0.0488</td> <td>    0.009</td> <td>   -5.344</td> <td> 0.000</td> <td>   -0.067</td> <td>   -0.031</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.3]</th>         <td>    0.3399</td> <td>    0.009</td> <td>   37.205</td> <td> 0.000</td> <td>    0.322</td> <td>    0.358</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.10]</th>        <td>   -0.2383</td> <td>    0.009</td> <td>  -25.343</td> <td> 0.000</td> <td>   -0.257</td> <td>   -0.220</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.11]</th>        <td>   -0.0477</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.15]</th>        <td>    0.1514</td> <td>    0.004</td> <td>   37.823</td> <td> 0.000</td> <td>    0.144</td> <td>    0.159</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.17]</th>        <td>   -0.1372</td> <td>    0.004</td> <td>  -34.291</td> <td> 0.000</td> <td>   -0.145</td> <td>   -0.129</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.18]</th>        <td>   -0.4369</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.19]</th>        <td>    0.5379</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.23]</th>        <td>    0.2981</td> <td>    0.004</td> <td>   74.504</td> <td> 0.000</td> <td>    0.290</td> <td>    0.306</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.25]</th>        <td>    0.1132</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.26]</th>        <td>    0.3175</td> <td>    0.008</td> <td>   38.385</td> <td> 0.000</td> <td>    0.301</td> <td>    0.334</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.27]</th>        <td>   -0.5922</td> <td>    0.023</td> <td>  -26.088</td> <td> 0.000</td> <td>   -0.637</td> <td>   -0.548</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.35]</th>        <td>   -1.3143</td> <td>    0.008</td> <td> -158.911</td> <td> 0.000</td> <td>   -1.330</td> <td>   -1.298</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.36]</th>        <td>   -0.0370</td> <td>    0.024</td> <td>   -1.549</td> <td> 0.121</td> <td>   -0.084</td> <td>    0.010</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.37]</th>        <td>   -0.2567</td> <td>      nan</td> <td>      nan</td> <td>   nan</td> <td>      nan</td> <td>      nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.41]</th>        <td>    0.0203</td> <td> 3.68e-10</td> <td> 5.51e+07</td> <td> 0.000</td> <td>    0.020</td> <td>    0.020</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.42]</th>        <td>   -0.7514</td> <td>    0.004</td> <td> -187.757</td> <td> 0.000</td> <td>   -0.759</td> <td>   -0.744</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.43]</th>        <td>   -0.0437</td> <td>    0.012</td> <td>   -3.617</td> <td> 0.000</td> <td>   -0.067</td> <td>   -0.020</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.44]</th>        <td>   -0.1892</td> <td>    0.012</td> <td>  -15.648</td> <td> 0.000</td> <td>   -0.213</td> <td>   -0.166</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(sid)[T.49]</th>        <td>   -0.6897</td> <td>    0.027</td> <td>  -25.257</td> <td> 0.000</td> <td>   -0.743</td> <td>   -0.636</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20023</th>              <td>    0.3523</td> <td>    0.060</td> <td>    5.829</td> <td> 0.000</td> <td>    0.234</td> <td>    0.471</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20072</th>              <td>   -0.0530</td> <td>    0.122</td> <td>   -0.436</td> <td> 0.663</td> <td>   -0.291</td> <td>    0.185</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20002</th>              <td>   -0.1470</td> <td>    0.238</td> <td>   -0.619</td> <td> 0.536</td> <td>   -0.613</td> <td>    0.319</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20081</th>              <td> -1.35e-16</td> <td> 1.16e-17</td> <td>  -11.634</td> <td> 0.000</td> <td>-1.58e-16</td> <td>-1.12e-16</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20001</th>              <td> 2.832e-16</td> <td> 1.46e-18</td> <td>  193.740</td> <td> 0.000</td> <td>  2.8e-16</td> <td> 2.86e-16</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20103</th>              <td>    0.2425</td> <td>    0.040</td> <td>    6.034</td> <td> 0.000</td> <td>    0.164</td> <td>    0.321</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20024</th>              <td>   -0.0757</td> <td>    0.101</td> <td>   -0.750</td> <td> 0.453</td> <td>   -0.274</td> <td>    0.122</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20034</th>              <td>    0.0622</td> <td>    0.056</td> <td>    1.105</td> <td> 0.269</td> <td>   -0.048</td> <td>    0.172</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20073</th>              <td>    0.2820</td> <td>    0.045</td> <td>    6.298</td> <td> 0.000</td> <td>    0.194</td> <td>    0.370</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20071</th>              <td>-2.165e-17</td> <td> 4.05e-31</td> <td>-5.35e+13</td> <td> 0.000</td> <td>-2.16e-17</td> <td>-2.16e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20031</th>              <td> 9.475e-18</td> <td> 6.53e-31</td> <td> 1.45e+13</td> <td> 0.000</td> <td> 9.48e-18</td> <td> 9.48e-18</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20094</th>              <td>   -0.1644</td> <td>    0.111</td> <td>   -1.476</td> <td> 0.140</td> <td>   -0.383</td> <td>    0.054</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20054</th>              <td>   -0.1900</td> <td>    0.108</td> <td>   -1.760</td> <td> 0.078</td> <td>   -0.402</td> <td>    0.022</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20013</th>              <td>    0.2653</td> <td>    0.076</td> <td>    3.474</td> <td> 0.001</td> <td>    0.116</td> <td>    0.415</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20061</th>              <td>-3.337e-17</td> <td> 1.85e-31</td> <td> -1.8e+14</td> <td> 0.000</td> <td>-3.34e-17</td> <td>-3.34e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20063</th>              <td>    0.2406</td> <td>    0.045</td> <td>    5.405</td> <td> 0.000</td> <td>    0.153</td> <td>    0.328</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20051</th>              <td>-4.331e-17</td> <td> 4.02e-31</td> <td>-1.08e+14</td> <td> 0.000</td> <td>-4.33e-17</td> <td>-4.33e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20101</th>              <td>-4.983e-17</td> <td> 9.07e-31</td> <td> -5.5e+13</td> <td> 0.000</td> <td>-4.98e-17</td> <td>-4.98e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20082</th>              <td>   -0.1451</td> <td>    0.136</td> <td>   -1.067</td> <td> 0.286</td> <td>   -0.411</td> <td>    0.121</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20033</th>              <td>    0.2976</td> <td>    0.045</td> <td>    6.546</td> <td> 0.000</td> <td>    0.209</td> <td>    0.387</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20014</th>              <td>    0.1652</td> <td>    0.105</td> <td>    1.568</td> <td> 0.117</td> <td>   -0.041</td> <td>    0.372</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20022</th>              <td>   -0.2325</td> <td>    0.091</td> <td>   -2.548</td> <td> 0.011</td> <td>   -0.411</td> <td>   -0.054</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20052</th>              <td>   -0.0145</td> <td>    0.071</td> <td>   -0.204</td> <td> 0.838</td> <td>   -0.154</td> <td>    0.125</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20044</th>              <td>    0.0679</td> <td>    0.047</td> <td>    1.433</td> <td> 0.152</td> <td>   -0.025</td> <td>    0.161</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20092</th>              <td>   -0.0148</td> <td>    0.089</td> <td>   -0.166</td> <td> 0.868</td> <td>   -0.190</td> <td>    0.160</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20004</th>              <td>   -0.0174</td> <td>    0.045</td> <td>   -0.391</td> <td> 0.696</td> <td>   -0.105</td> <td>    0.070</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20053</th>              <td>    0.3158</td> <td>    0.048</td> <td>    6.613</td> <td> 0.000</td> <td>    0.222</td> <td>    0.409</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20043</th>              <td>    0.2405</td> <td>    0.038</td> <td>    6.255</td> <td> 0.000</td> <td>    0.165</td> <td>    0.316</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20021</th>              <td>         0</td> <td>        0</td> <td>      nan</td> <td>   nan</td> <td>        0</td> <td>        0</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20003</th>              <td>    0.4246</td> <td>    0.062</td> <td>    6.824</td> <td> 0.000</td> <td>    0.303</td> <td>    0.547</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20104</th>              <td>   -0.1102</td> <td>    0.038</td> <td>   -2.935</td> <td> 0.003</td> <td>   -0.184</td> <td>   -0.037</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20032</th>              <td>   -0.2329</td> <td>    0.078</td> <td>   -2.970</td> <td> 0.003</td> <td>   -0.387</td> <td>   -0.079</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20074</th>              <td>   -0.0593</td> <td>    0.116</td> <td>   -0.509</td> <td> 0.610</td> <td>   -0.288</td> <td>    0.169</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20102</th>              <td>   -0.0625</td> <td>    0.060</td> <td>   -1.044</td> <td> 0.296</td> <td>   -0.180</td> <td>    0.055</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20011</th>              <td>         0</td> <td>        0</td> <td>      nan</td> <td>   nan</td> <td>        0</td> <td>        0</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20041</th>              <td>         0</td> <td>        0</td> <td>      nan</td> <td>   nan</td> <td>        0</td> <td>        0</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20091</th>              <td>         0</td> <td>        0</td> <td>      nan</td> <td>   nan</td> <td>        0</td> <td>        0</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20012</th>              <td>   -0.2904</td> <td>    0.124</td> <td>   -2.344</td> <td> 0.019</td> <td>   -0.533</td> <td>   -0.048</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20093</th>              <td>    0.2871</td> <td>    0.056</td> <td>    5.113</td> <td> 0.000</td> <td>    0.177</td> <td>    0.397</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20062</th>              <td>   -0.1065</td> <td>    0.075</td> <td>   -1.426</td> <td> 0.154</td> <td>   -0.253</td> <td>    0.040</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20084</th>              <td>   -0.0138</td> <td>    0.123</td> <td>   -0.112</td> <td> 0.910</td> <td>   -0.255</td> <td>    0.227</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20083</th>              <td>    0.2591</td> <td>    0.061</td> <td>    4.267</td> <td> 0.000</td> <td>    0.140</td> <td>    0.378</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20064</th>              <td>    0.0346</td> <td>    0.059</td> <td>    0.589</td> <td> 0.556</td> <td>   -0.081</td> <td>    0.150</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>r20042</th>              <td>   -0.1737</td> <td>    0.036</td> <td>   -4.846</td> <td> 0.000</td> <td>   -0.244</td> <td>   -0.103</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>35.847</td> <th>  Durbin-Watson:     </th> <td>   1.356</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 201.726</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.367</td> <th>  Prob(JB):          </th> <td>1.57e-44</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 7.519</td> <th>  Cond. No.          </th> <td>6.48e+16</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 6.86e-32. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:             l_homicide   R-squared:                       0.900\n",
+       "Model:                            OLS   Adj. R-squared:                  0.862\n",
+       "Method:                 Least Squares   F-statistic:                 1.369e+13\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):          3.99e-127\n",
+       "Time:                        13:30:29   Log-Likelihood:                 88.377\n",
+       "No. Observations:                 231   AIC:                            -48.75\n",
+       "Df Residuals:                     167   BIC:                             171.6\n",
+       "Df Model:                          63                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "=======================================================================================\n",
+       "                          coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "---------------------------------------------------------------------------------------\n",
+       "Intercept               1.4335      0.028     50.373      0.000       1.378       1.489\n",
+       "C(time_til)[T.-8.0]     0.0251      0.225      0.112      0.911      -0.417       0.467\n",
+       "C(time_til)[T.-7.0]    -0.0308      0.172     -0.179      0.858      -0.368       0.306\n",
+       "C(time_til)[T.-6.0]     0.2006      0.076      2.630      0.009       0.051       0.350\n",
+       "C(time_til)[T.-5.0]     0.1677      0.040      4.187      0.000       0.089       0.246\n",
+       "C(time_til)[T.-4.0]     0.1493      0.072      2.064      0.039       0.008       0.291\n",
+       "C(time_til)[T.-3.0]     0.1816      0.051      3.578      0.000       0.082       0.281\n",
+       "C(time_til)[T.-2.0]     0.1777      0.059      3.001      0.003       0.062       0.294\n",
+       "C(time_til)[T.-1.0]     0.0910      0.059      1.532      0.125      -0.025       0.207\n",
+       "C(time_til)[T.0.0]      0.2041      0.081      2.509      0.012       0.045       0.364\n",
+       "C(time_til)[T.1.0]      0.1575      0.060      2.645      0.008       0.041       0.274\n",
+       "C(time_til)[T.2.0]      0.2202      0.065      3.391      0.001       0.093       0.347\n",
+       "C(time_til)[T.3.0]      0.1171      0.065      1.812      0.070      -0.010       0.244\n",
+       "C(time_til)[T.4.0]      0.0682      0.083      0.823      0.410      -0.094       0.230\n",
+       "C(time_til)[T.5.0]      0.1440      0.053      2.738      0.006       0.041       0.247\n",
+       "C(year)[T.2001]         0.1401      0.058      2.413      0.016       0.026       0.254\n",
+       "C(year)[T.2002]         0.0441      0.037      1.179      0.238      -0.029       0.118\n",
+       "C(year)[T.2003]         0.1269      0.026      4.832      0.000       0.075       0.178\n",
+       "C(year)[T.2004]         0.1348      0.042      3.191      0.001       0.052       0.218\n",
+       "C(year)[T.2005]         0.1113      0.063      1.754      0.079      -0.013       0.236\n",
+       "C(year)[T.2006]         0.1687      0.058      2.927      0.003       0.056       0.282\n",
+       "C(year)[T.2007]         0.1696      0.069      2.460      0.014       0.035       0.305\n",
+       "C(year)[T.2008]         0.1001      0.055      1.810      0.070      -0.008       0.209\n",
+       "C(year)[T.2009]         0.1079      0.052      2.083      0.037       0.006       0.209\n",
+       "C(year)[T.2010]         0.0697      0.044      1.596      0.110      -0.016       0.155\n",
+       "C(sid)[T.2]            -0.0488      0.009     -5.344      0.000      -0.067      -0.031\n",
+       "C(sid)[T.3]             0.3399      0.009     37.205      0.000       0.322       0.358\n",
+       "C(sid)[T.10]           -0.2383      0.009    -25.343      0.000      -0.257      -0.220\n",
+       "C(sid)[T.11]           -0.0477        nan        nan        nan         nan         nan\n",
+       "C(sid)[T.15]            0.1514      0.004     37.823      0.000       0.144       0.159\n",
+       "C(sid)[T.17]           -0.1372      0.004    -34.291      0.000      -0.145      -0.129\n",
+       "C(sid)[T.18]           -0.4369        nan        nan        nan         nan         nan\n",
+       "C(sid)[T.19]            0.5379        nan        nan        nan         nan         nan\n",
+       "C(sid)[T.23]            0.2981      0.004     74.504      0.000       0.290       0.306\n",
+       "C(sid)[T.25]            0.1132        nan        nan        nan         nan         nan\n",
+       "C(sid)[T.26]            0.3175      0.008     38.385      0.000       0.301       0.334\n",
+       "C(sid)[T.27]           -0.5922      0.023    -26.088      0.000      -0.637      -0.548\n",
+       "C(sid)[T.35]           -1.3143      0.008   -158.911      0.000      -1.330      -1.298\n",
+       "C(sid)[T.36]           -0.0370      0.024     -1.549      0.121      -0.084       0.010\n",
+       "C(sid)[T.37]           -0.2567        nan        nan        nan         nan         nan\n",
+       "C(sid)[T.41]            0.0203   3.68e-10   5.51e+07      0.000       0.020       0.020\n",
+       "C(sid)[T.42]           -0.7514      0.004   -187.757      0.000      -0.759      -0.744\n",
+       "C(sid)[T.43]           -0.0437      0.012     -3.617      0.000      -0.067      -0.020\n",
+       "C(sid)[T.44]           -0.1892      0.012    -15.648      0.000      -0.213      -0.166\n",
+       "C(sid)[T.49]           -0.6897      0.027    -25.257      0.000      -0.743      -0.636\n",
+       "r20023                  0.3523      0.060      5.829      0.000       0.234       0.471\n",
+       "r20072                 -0.0530      0.122     -0.436      0.663      -0.291       0.185\n",
+       "r20002                 -0.1470      0.238     -0.619      0.536      -0.613       0.319\n",
+       "r20081               -1.35e-16   1.16e-17    -11.634      0.000   -1.58e-16   -1.12e-16\n",
+       "r20001               2.832e-16   1.46e-18    193.740      0.000     2.8e-16    2.86e-16\n",
+       "r20103                  0.2425      0.040      6.034      0.000       0.164       0.321\n",
+       "r20024                 -0.0757      0.101     -0.750      0.453      -0.274       0.122\n",
+       "r20034                  0.0622      0.056      1.105      0.269      -0.048       0.172\n",
+       "r20073                  0.2820      0.045      6.298      0.000       0.194       0.370\n",
+       "r20071              -2.165e-17   4.05e-31  -5.35e+13      0.000   -2.16e-17   -2.16e-17\n",
+       "r20031               9.475e-18   6.53e-31   1.45e+13      0.000    9.48e-18    9.48e-18\n",
+       "r20094                 -0.1644      0.111     -1.476      0.140      -0.383       0.054\n",
+       "r20054                 -0.1900      0.108     -1.760      0.078      -0.402       0.022\n",
+       "r20013                  0.2653      0.076      3.474      0.001       0.116       0.415\n",
+       "r20061              -3.337e-17   1.85e-31   -1.8e+14      0.000   -3.34e-17   -3.34e-17\n",
+       "r20063                  0.2406      0.045      5.405      0.000       0.153       0.328\n",
+       "r20051              -4.331e-17   4.02e-31  -1.08e+14      0.000   -4.33e-17   -4.33e-17\n",
+       "r20101              -4.983e-17   9.07e-31   -5.5e+13      0.000   -4.98e-17   -4.98e-17\n",
+       "r20082                 -0.1451      0.136     -1.067      0.286      -0.411       0.121\n",
+       "r20033                  0.2976      0.045      6.546      0.000       0.209       0.387\n",
+       "r20014                  0.1652      0.105      1.568      0.117      -0.041       0.372\n",
+       "r20022                 -0.2325      0.091     -2.548      0.011      -0.411      -0.054\n",
+       "r20052                 -0.0145      0.071     -0.204      0.838      -0.154       0.125\n",
+       "r20044                  0.0679      0.047      1.433      0.152      -0.025       0.161\n",
+       "r20092                 -0.0148      0.089     -0.166      0.868      -0.190       0.160\n",
+       "r20004                 -0.0174      0.045     -0.391      0.696      -0.105       0.070\n",
+       "r20053                  0.3158      0.048      6.613      0.000       0.222       0.409\n",
+       "r20043                  0.2405      0.038      6.255      0.000       0.165       0.316\n",
+       "r20021                       0          0        nan        nan           0           0\n",
+       "r20003                  0.4246      0.062      6.824      0.000       0.303       0.547\n",
+       "r20104                 -0.1102      0.038     -2.935      0.003      -0.184      -0.037\n",
+       "r20032                 -0.2329      0.078     -2.970      0.003      -0.387      -0.079\n",
+       "r20074                 -0.0593      0.116     -0.509      0.610      -0.288       0.169\n",
+       "r20102                 -0.0625      0.060     -1.044      0.296      -0.180       0.055\n",
+       "r20011                       0          0        nan        nan           0           0\n",
+       "r20041                       0          0        nan        nan           0           0\n",
+       "r20091                       0          0        nan        nan           0           0\n",
+       "r20012                 -0.2904      0.124     -2.344      0.019      -0.533      -0.048\n",
+       "r20093                  0.2871      0.056      5.113      0.000       0.177       0.397\n",
+       "r20062                 -0.1065      0.075     -1.426      0.154      -0.253       0.040\n",
+       "r20084                 -0.0138      0.123     -0.112      0.910      -0.255       0.227\n",
+       "r20083                  0.2591      0.061      4.267      0.000       0.140       0.378\n",
+       "r20064                  0.0346      0.059      0.589      0.556      -0.081       0.150\n",
+       "r20042                 -0.1737      0.036     -4.846      0.000      -0.244      -0.103\n",
+       "==============================================================================\n",
+       "Omnibus:                       35.847   Durbin-Watson:                   1.356\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              201.726\n",
+       "Skew:                          -0.367   Prob(JB):                     1.57e-44\n",
+       "Kurtosis:                       7.519   Cond. No.                     6.48e+16\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 6.86e-32. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "castle['time_til'] = castle['year'] - castle['treatment_date']\n",
+    "castle = castle[~pd.isnull(castle.time_til)]\n",
+    "formula = \"l_homicide ~ {} + C(time_til) + C(year) + C(sid)\".format(\"+\".join(region))\n",
+    "\n",
+    "event_study_formula = sm.OLS.from_formula(formula,\n",
+    "            data = castle, \n",
+    "            freq_weights = castle['popwt']).fit(cov_type='cluster', cov_kwds={'groups':castle['sid']})\n",
+    "event_study_formula.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8768291253333)>"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# grab the clustered standard errors\n",
+    "# and average coefficient estimates\n",
+    "# from the regression, label them accordingly\n",
+    "# add a zero'th lag for plotting purposes\n",
+    "\n",
+    "leadslags_plot = pd.DataFrame({\n",
+    "    'sd' : np.diag(event_study_formula.cov_params().values)[1:15],\n",
+    "    'mean':  event_study_formula.params[1:15],\n",
+    "    'label': np.arange(-9, 5)}) \n",
+    "\n",
+    "leadslags_plot['lb'] = leadslags_plot['mean'] - leadslags_plot['sd']*1.96\n",
+    "leadslags_plot['ub'] = leadslags_plot['mean'] + leadslags_plot['sd']*1.96\n",
+    "\n",
+    "# This version has a point-range at each\n",
+    "# estimated lead or lag\n",
+    "# comes down to stylistic preference at the\n",
+    "# end of the day!\n",
+    "p.ggplot(leadslags_plot, p.aes(x = 'label', y = 'mean',\n",
+    "             ymin = 'lb', \n",
+    "             ymax = 'ub')) +\\\n",
+    "    p.geom_hline(yintercept = 0.035169444, color = \"red\") +\\\n",
+    "    p.geom_pointrange() +\\\n",
+    "    p.theme_minimal() +\\\n",
+    "    p.xlab(\"Years before and after castle doctrine expansion\") +\\\n",
+    "    p.ylab(\"log(Homicide Rate)\") +\\\n",
+    "    p.geom_hline(yintercept = 0,\n",
+    "             linetype = \"dashed\") +\\\n",
+    "    p.geom_vline(xintercept = 0,\n",
+    "             linetype = \"dashed\")\n",
+    "  "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTIONS\n",
+    "- Put into your own words why we estimated the pre-treatment leads?\n",
+    "- Put into your own words what we expected to find?\n",
+    "- How convinced are you by this analysis that parallel trends was likely to hold in Cheng and Hoekstra's data? \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Python/Directed_Acyclical_Graphs.ipynb b/Python/Directed_Acyclical_Graphs.ipynb
new file mode 100644
index 0000000..fe14c05
--- /dev/null
+++ b/Python/Directed_Acyclical_Graphs.ipynb
@@ -0,0 +1,353 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Directed Acyclical Graphs** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import statsmodels.api as sm\n",
+    "from stargazer.stargazer import Stargazer\n",
+    "\n",
+    "import plotnine as p"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_csv(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "## Collider - Discrimination"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "tb = pd.DataFrame({\n",
+    "    'female': np.random.binomial(1, .5, size=10000),\n",
+    "    'ability': np.random.normal(size=10000)})\n",
+    "tb['discrimination'] = tb.female.copy()\n",
+    "tb['occupation'] = 1 + 2*tb['ability'] + 0*tb['female'] - 2*tb['discrimination'] + np.random.normal(size=10000)\n",
+    "tb['wage'] = 1 - 1*tb['discrimination'] + 1*tb['occupation'] + 2*tb['ability'] + np.random.normal(size=10000) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><td colspan=\"3\"><em>Dependent variable:wage</em></td></tr><tr><td style=\"text-align:left\"></td><tr><td></td><td colspan=\"1\">Biased Unconditional</td><td colspan=\"1\">Biased</td><td colspan=\"1\">Unbiased Conditional</td></tr><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>1.973<sup>***</sup></td><td>0.177<sup>***</sup></td><td>0.974<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.060)</td><td>(0.020)</td><td>(0.017)</td></tr><tr><td style=\"text-align:left\">ability</td><td></td><td></td><td>1.954<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td></td><td>(0.022)</td></tr><tr><td style=\"text-align:left\">female</td><td>-2.945<sup>***</sup></td><td>0.647<sup>***</sup></td><td>-0.932<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.085)</td><td>(0.029)</td><td>(0.028)</td></tr><tr><td style=\"text-align:left\">occupation</td><td></td><td>1.805<sup>***</sup></td><td>1.026<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.006)</td><td>(0.010)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>10,000</td><td>10,000</td><td>10,000</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.108</td><td>0.913</td><td>0.951</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.108</td><td>0.913</td><td>0.951</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>4.233 (df=9998)</td><td>1.321 (df=9997)</td><td>0.993 (df=9996)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>1209.615<sup>***</sup> (df=1; 9998)</td><td>52561.170<sup>***</sup> (df=2; 9997)</td><td>64572.581<sup>***</sup> (df=3; 9996)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fb1d04dae80>"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "lm_1 = sm.OLS.from_formula('wage ~ female', data=tb).fit()\n",
+    "lm_2 = sm.OLS.from_formula('wage ~ female + occupation', data=tb).fit()\n",
+    "lm_3 = sm.OLS.from_formula('wage ~ female + occupation + ability', data=tb).fit()\n",
+    "\n",
+    "\n",
+    "st = Stargazer((lm_1,lm_2,lm_3))\n",
+    "st.custom_columns([\"Biased Unconditional\", \"Biased\", \"Unbiased Conditional\"], [1, 1, 1])\n",
+    "st"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTIONS\n",
+    "- What is the true direct effect of discrimination on wages?  \n",
+    "- Explain the channels by which discrimination impacts wages.  \n",
+    "- What makes occupation a collider?\n",
+    "- What controls are necessary to eliminate this collider bias?\n",
+    "\n",
+    "\n",
+    "\n",
+    "## Movie Star"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>beauty</th>\n",
+       "      <th>talent</th>\n",
+       "      <th>score</th>\n",
+       "      <th>c85</th>\n",
+       "      <th>star</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>0.418926</td>\n",
+       "      <td>1.031041</td>\n",
+       "      <td>1.449967</td>\n",
+       "      <td>1.396979</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>0.244585</td>\n",
+       "      <td>-3.653642</td>\n",
+       "      <td>-3.409057</td>\n",
+       "      <td>1.396979</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>-0.262557</td>\n",
+       "      <td>-1.550817</td>\n",
+       "      <td>-1.813374</td>\n",
+       "      <td>1.396979</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>-0.082991</td>\n",
+       "      <td>-0.334602</td>\n",
+       "      <td>-0.417593</td>\n",
+       "      <td>1.396979</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>-0.125137</td>\n",
+       "      <td>-2.227503</td>\n",
+       "      <td>-2.352640</td>\n",
+       "      <td>1.396979</td>\n",
+       "      <td>0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     beauty    talent     score       c85  star\n",
+       "0  0.418926  1.031041  1.449967  1.396979     1\n",
+       "1  0.244585 -3.653642 -3.409057  1.396979     0\n",
+       "2 -0.262557 -1.550817 -1.813374  1.396979     0\n",
+       "3 -0.082991 -0.334602 -0.417593  1.396979     0\n",
+       "4 -0.125137 -2.227503 -2.352640  1.396979     0"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "start_is_born = pd.DataFrame({\n",
+    "    'beauty': np.random.normal(size=2500),\n",
+    "    'talent': np.random.normal(size=2500)})\n",
+    "    \n",
+    "start_is_born['score'] = start_is_born['beauty'] + start_is_born['talent']\n",
+    "start_is_born['c85'] = np.percentile(start_is_born['score'], q=85)\n",
+    "start_is_born['star'] = 0\n",
+    "start_is_born.loc[start_is_born['score']>start_is_born['c85'], 'star'] = 1\n",
+    "start_is_born.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8775105510946)>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm = sm.OLS.from_formula('beauty ~ talent', data=start_is_born).fit()\n",
+    "\n",
+    "p.ggplot(start_is_born, p.aes(x='talent', y='beauty')) +\\\n",
+    "    p.geom_point(size = 0.5) +\\\n",
+    "    p.xlim(-4, 4) +\\\n",
+    "    p.ylim(-4, 4)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8775105528971)>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(start_is_born[start_is_born.star==1], p.aes(x='talent', y='beauty')) +\\\n",
+    "    p.geom_point(size = 0.5) +\\\n",
+    "    p.xlim(-4, 4) +\\\n",
+    "    p.ylim(-4, 4)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8775106119870)>"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(start_is_born[start_is_born.star==0], p.aes(x='talent', y='beauty')) +\\\n",
+    "    p.geom_point(size = 0.5) +\\\n",
+    "    p.xlim(-4, 4) +\\\n",
+    "    p.ylim(-4, 4)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTIONS\n",
+    "- What is the correlation between talent and beauty among stars?  Non-stars?\n",
+    "- But what is the correlation between talent and beauty in the population?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Python/Instrumental_Variables.ipynb b/Python/Instrumental_Variables.ipynb
new file mode 100644
index 0000000..8b805eb
--- /dev/null
+++ b/Python/Instrumental_Variables.ipynb
@@ -0,0 +1,959 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Instrumental Variables** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n",
+    "\n",
+    "### Packages needed\n",
+    "\n",
+    "The first thing you need to do is install a few packages to make sure everything runs:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import plotnine as p\n",
+    "\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "from linearmodels import IV2SLS \n",
+    "\n",
+    "from stargazer.stargazer import Stargazer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def read_data(file):\n",
+    "    full_path = \"https://raw.github.com/scunning1975/mixtape/master/\" + file\n",
+    "    \n",
+    "    return pd.read_stata(full_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def lm_robust(formula, data, group_col):\n",
+    "    regression = sm.OLS.from_formula(formula, data = data)\n",
+    "    regression = regression.fit(cov_type=\"cluster\",cov_kwds={\"groups\":data[group_col]})\n",
+    "    return regression"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Card"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>          <td>lwage</td>      <th>  R-squared:         </th> <td>   0.305</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.304</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   219.2</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>1.97e-232</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:31:12</td>     <th>  Log-Likelihood:    </th> <td> -1273.9</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  3003</td>      <th>  AIC:               </th> <td>   2562.</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>  2996</td>      <th>  BIC:               </th> <td>   2604.</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     6</td>      <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th> <td>    5.0633</td> <td>    0.064</td> <td>   79.437</td> <td> 0.000</td> <td>    4.938</td> <td>    5.188</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>educ</th>      <td>    0.0712</td> <td>    0.003</td> <td>   20.438</td> <td> 0.000</td> <td>    0.064</td> <td>    0.078</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>exper</th>     <td>    0.0342</td> <td>    0.002</td> <td>   15.422</td> <td> 0.000</td> <td>    0.030</td> <td>    0.038</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black</th>     <td>   -0.1660</td> <td>    0.018</td> <td>   -9.426</td> <td> 0.000</td> <td>   -0.201</td> <td>   -0.131</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>south</th>     <td>   -0.1316</td> <td>    0.015</td> <td>   -8.788</td> <td> 0.000</td> <td>   -0.161</td> <td>   -0.102</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>married</th>   <td>   -0.0359</td> <td>    0.003</td> <td>  -10.547</td> <td> 0.000</td> <td>   -0.043</td> <td>   -0.029</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>smsa</th>      <td>    0.1758</td> <td>    0.015</td> <td>   11.372</td> <td> 0.000</td> <td>    0.145</td> <td>    0.206</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>53.196</td> <th>  Durbin-Watson:     </th> <td>   1.858</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td>  69.430</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.231</td> <th>  Prob(JB):          </th> <td>8.38e-16</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 3.584</td> <th>  Cond. No.          </th> <td>    154.</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                  lwage   R-squared:                       0.305\n",
+       "Model:                            OLS   Adj. R-squared:                  0.304\n",
+       "Method:                 Least Squares   F-statistic:                     219.2\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):          1.97e-232\n",
+       "Time:                        13:31:12   Log-Likelihood:                -1273.9\n",
+       "No. Observations:                3003   AIC:                             2562.\n",
+       "Df Residuals:                    2996   BIC:                             2604.\n",
+       "Df Model:                           6                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "==============================================================================\n",
+       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept      5.0633      0.064     79.437      0.000       4.938       5.188\n",
+       "educ           0.0712      0.003     20.438      0.000       0.064       0.078\n",
+       "exper          0.0342      0.002     15.422      0.000       0.030       0.038\n",
+       "black         -0.1660      0.018     -9.426      0.000      -0.201      -0.131\n",
+       "south         -0.1316      0.015     -8.788      0.000      -0.161      -0.102\n",
+       "married       -0.0359      0.003    -10.547      0.000      -0.043      -0.029\n",
+       "smsa           0.1758      0.015     11.372      0.000       0.145       0.206\n",
+       "==============================================================================\n",
+       "Omnibus:                       53.196   Durbin-Watson:                   1.858\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               69.430\n",
+       "Skew:                          -0.231   Prob(JB):                     8.38e-16\n",
+       "Kurtosis:                       3.584   Cond. No.                         154.\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#card = read_data(\"card.dta\")\n",
+    "card = read_data(\"card.dta\")\n",
+    "\n",
+    "#OLS\n",
+    "ols_reg = sm.OLS.from_formula(\"lwage ~ educ + exper + black + south + married + smsa\", \n",
+    "              data = card).fit()\n",
+    "\n",
+    "ols_reg.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/linearmodels/shared/exceptions.py:35: MissingValueWarning: \n",
+      "Inputs contain missing values. Dropping rows with missing observations.\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>IV-2SLS Estimation Summary</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>          <td>lwage</td>      <th>  R-squared:         </th> <td>0.2513</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Estimator:</th>             <td>IV-2SLS</td>     <th>  Adj. R-squared:    </th> <td>0.2498</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>       <td>3003</td>       <th>  F-statistic:       </th> <td>892.71</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, Mar 07 2021</td> <th>  P-value (F-stat)   </th> <td>0.0000</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:31:12</td>     <th>  Distribution:      </th> <td>chi2(6)</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Cov. Estimator:</th>        <td>robust</td>      <th>                     </th>    <td></td>    \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th></th>                          <td></td>         <th>                     </th>    <td></td>    \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<caption>Parameter Estimates</caption>\n",
+       "<tr>\n",
+       "      <td></td>      <th>Parameter</th> <th>Std. Err.</th> <th>T-stat</th>  <th>P-value</th> <th>Lower CI</th> <th>Upper CI</th>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>  <td>4.1625</td>    <td>0.8349</td>   <td>4.9857</td>  <td>0.0000</td>   <td>2.5262</td>   <td>5.7988</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>exper</th>      <td>0.0556</td>    <td>0.0199</td>   <td>2.7980</td>  <td>0.0051</td>   <td>0.0166</td>   <td>0.0945</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black</th>      <td>-0.1157</td>   <td>0.0496</td>   <td>-2.3343</td> <td>0.0196</td>   <td>-0.2128</td>  <td>-0.0186</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>south</th>      <td>-0.1132</td>   <td>0.0229</td>   <td>-4.9314</td> <td>0.0000</td>   <td>-0.1581</td>  <td>-0.0682</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>married</th>    <td>-0.0320</td>   <td>0.0051</td>   <td>-6.3037</td> <td>0.0000</td>   <td>-0.0419</td>  <td>-0.0220</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>smsa</th>       <td>0.1477</td>    <td>0.0303</td>   <td>4.8721</td>  <td>0.0000</td>   <td>0.0883</td>   <td>0.2071</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>educ</th>       <td>0.1242</td>    <td>0.0492</td>   <td>2.5258</td>  <td>0.0115</td>   <td>0.0278</td>   <td>0.2205</td> \n",
+       "</tr>\n",
+       "</table><br/><br/>Endogenous: educ<br/>Instruments: nearc4<br/>Robust Covariance (Heteroskedastic)<br/>Debiased: False"
+      ],
+      "text/plain": [
+       "<class 'linearmodels.compat.statsmodels.Summary'>\n",
+       "\"\"\"\n",
+       "                          IV-2SLS Estimation Summary                          \n",
+       "==============================================================================\n",
+       "Dep. Variable:                  lwage   R-squared:                      0.2513\n",
+       "Estimator:                    IV-2SLS   Adj. R-squared:                 0.2498\n",
+       "No. Observations:                3003   F-statistic:                    892.71\n",
+       "Date:                Sun, Mar 07 2021   P-value (F-stat)                0.0000\n",
+       "Time:                        13:31:12   Distribution:                  chi2(6)\n",
+       "Cov. Estimator:                robust                                         \n",
+       "                                                                              \n",
+       "                             Parameter Estimates                              \n",
+       "==============================================================================\n",
+       "            Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept      4.1625     0.8349     4.9857     0.0000      2.5262      5.7988\n",
+       "exper          0.0556     0.0199     2.7980     0.0051      0.0166      0.0945\n",
+       "black         -0.1157     0.0496    -2.3343     0.0196     -0.2128     -0.0186\n",
+       "south         -0.1132     0.0229    -4.9314     0.0000     -0.1581     -0.0682\n",
+       "married       -0.0320     0.0051    -6.3037     0.0000     -0.0419     -0.0220\n",
+       "smsa           0.1477     0.0303     4.8721     0.0000      0.0883      0.2071\n",
+       "educ           0.1242     0.0492     2.5258     0.0115      0.0278      0.2205\n",
+       "==============================================================================\n",
+       "\n",
+       "Endogenous: educ\n",
+       "Instruments: nearc4\n",
+       "Robust Covariance (Heteroskedastic)\n",
+       "Debiased: False\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#2SLS\n",
+    "iv_reg = IV2SLS.from_formula(\"lwage ~ 1 + exper + black + south + married + smsa + [educ ~ nearc4 ]\", card).fit()\n",
+    "iv_reg.summary"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Interpret the coefficient on education when we used OLS versus when used 2SLS. \n",
+    "- How does the estimated effect of education change when instrumenting with being close to a 4-year college?  That is, does the coefficient get larger or smaller compared to OLS?\n",
+    "- If the only source of bias in our OLS regression was omitted heterogeneous ability, then will 2SLS be larger, smaller or the same as OLS estimate?  Why/why not?   \n",
+    "- Is the finding of the causal effect of educating when using 2SLS, when compared to the estimate using OLS, consistent with ability bias?  What else do you think may be going on and why?\n",
+    "- What sorts of individuals will go to college regardless of whether a college is near them?  What sorts of individuals will never go to a college even if one is near them?  And what sorts of people will go to a college if one is near them but won't go to college if it is not near them?\n",
+    "\n",
+    "## JIVE "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "OLS\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"3\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><td colspan=\"2\"><em>Dependent variable:guilt</em></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td></tr><tr><td colspan=\"3\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">DUI1st</td><td></td><td>0.047<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">F1</td><td></td><td>0.016<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">F2</td><td></td><td>0.052<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">F3</td><td></td><td>0.096<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">Intercept</td><td>-0.413<sup>***</sup></td><td>-0.549<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.128)</td><td>(0.121)</td></tr><tr><td style=\"text-align:left\">M</td><td></td><td>0.253<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">M1</td><td></td><td>0.025<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.002)</td></tr><tr><td style=\"text-align:left\">M2</td><td></td><td>-0.068<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">M3</td><td></td><td>0.133<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">age</td><td></td><td>0.001<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">aggAss</td><td></td><td>0.008<sup>**</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">bailDate</td><td>0.000<sup>***</sup></td><td>0.000<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.000)</td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">black</td><td></td><td>0.070<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">day</td><td>-0.000<sup>**</sup></td><td>-0.000<sup>**</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.000)</td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">day2</td><td>0.000<sup></sup></td><td>0.000<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.000)</td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">drugSell</td><td></td><td>0.047<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.005)</td></tr><tr><td style=\"text-align:left\">fel</td><td></td><td>-0.004<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">jail3</td><td>-0.001<sup></sup></td><td>0.029<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.002)</td><td>(0.002)</td></tr><tr><td style=\"text-align:left\">male</td><td></td><td>-0.036<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.002)</td></tr><tr><td style=\"text-align:left\">mis</td><td></td><td>0.112<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">onePrior</td><td></td><td>0.050<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">possess</td><td></td><td>-0.061<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">priorCases</td><td></td><td>-0.006<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">priorWI5</td><td></td><td>0.038<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><tr><td style=\"text-align:left\">prior_felChar</td><td></td><td>-0.006<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.000)</td></tr><tr><td style=\"text-align:left\">prior_guilt</td><td></td><td>0.026<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.001)</td></tr><tr><td style=\"text-align:left\">robbery</td><td></td><td>-0.081<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">sum</td><td></td><td>0.066<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">t1</td><td>0.006<sup></sup></td><td>-0.018<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.016)</td><td>(0.015)</td></tr><tr><td style=\"text-align:left\">t2</td><td>-0.017<sup></sup></td><td>-0.027<sup>**</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.012)</td><td>(0.012)</td></tr><tr><td style=\"text-align:left\">t3</td><td>0.000<sup></sup></td><td>-0.002<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.011)</td><td>(0.010)</td></tr><tr><td style=\"text-align:left\">t4</td><td>0.012<sup>*</sup></td><td>0.009<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.007)</td><td>(0.006)</td></tr><tr><td style=\"text-align:left\">t5</td><td>-0.005<sup></sup></td><td>0.002<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.004)</td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">threePriors</td><td></td><td>0.008<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.002)</td></tr><tr><td style=\"text-align:left\">white</td><td></td><td>0.107<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.003)</td></tr><td colspan=\"3\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>331,971</td><td>331,971</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.007</td><td>0.117</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.007</td><td>0.117</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>0.498 (df=331961)</td><td>0.470 (df=331936)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>276.514<sup>***</sup> (df=9; 331961)</td><td>1288.725<sup>***</sup> (df=34; 331936)</td></tr><tr><td colspan=\"3\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"2\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7ff8461b2c40>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "judge = read_data(\"judge_fe.dta\")\n",
+    "judge['bailDate'] = (judge['bailDate'] - pd.to_datetime('1970-01-01')).dt.days.values\n",
+    "\n",
+    "# grouped variable names from the data set\n",
+    "judge_pre = \"+\".join(judge.columns[judge.columns.str.contains('^judge_pre_[1-7]')])\n",
+    "demo = \"+\".join(['black', 'age', 'male', 'white'])\n",
+    "off = \"+\".join(['fel', 'mis', 'sum', 'F1', 'F2', 'F3', 'M1', 'M2', 'M3', 'M'])\n",
+    "prior = \"+\".join(['priorCases', 'priorWI5', 'prior_felChar', 'prior_guilt', 'onePrior', 'threePriors'])\n",
+    "control2 = \"+\".join(['day', 'day2', 'bailDate', 't1', 't2', 't3', 't4', 't5'])\n",
+    "\n",
+    "#formulas used in the OLS\n",
+    "min_formula = \"guilt ~ jail3 + \" + control2\n",
+    "max_formula = \"\"\"guilt ~ jail3 + possess + robbery + DUI1st + drugSell + \n",
+    "                aggAss + {demo} + {prior} + {off} + {control2}\"\"\".format(demo=demo,\n",
+    "                                                                        prior=prior,\n",
+    "                                                                        off=off,\n",
+    "                                                                        control2=control2)\n",
+    "\n",
+    "#max variables and min variables\n",
+    "min_ols = sm.OLS.from_formula(min_formula, data = judge).fit()\n",
+    "max_ols = sm.OLS.from_formula(max_formula, data = judge).fit()\n",
+    "print(\"OLS\")\n",
+    "Stargazer([min_ols, max_ols])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "IV\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>IV-2SLS Estimation Summary</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>          <td>guilt</td>      <th>  R-squared:         </th>  <td>0.4849</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Estimator:</th>             <td>IV-2SLS</td>     <th>  Adj. R-squared:    </th>  <td>0.4849</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>331971</td>      <th>  F-statistic:       </th> <td>3.199e+05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, Mar 07 2021</td> <th>  P-value (F-stat)   </th>  <td>0.0000</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:31:29</td>     <th>  Distribution:      </th>  <td>chi2(9)</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Cov. Estimator:</th>        <td>robust</td>      <th>                     </th>     <td></td>     \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th></th>                          <td></td>         <th>                     </th>     <td></td>     \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<caption>Parameter Estimates</caption>\n",
+       "<tr>\n",
+       "      <td></td>      <th>Parameter</th> <th>Std. Err.</th> <th>T-stat</th>  <th>P-value</th> <th>Lower CI</th>  <th>Upper CI</th> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>day</th>      <td>-7.952e-05</td> <td>2.506e-05</td> <td>-3.1730</td> <td>0.0015</td>   <td>-0.0001</td>  <td>-3.04e-05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>day2</th>      <td> 1.4e-07</td>  <td>5.806e-08</td> <td>2.4111</td>  <td>0.0159</td>  <td>2.619e-08</td> <td>2.538e-07</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>bailDate</th>  <td> 3.2e-05</td>  <td>1.902e-06</td> <td>16.829</td>  <td>0.0000</td>  <td>2.828e-05</td> <td>3.573e-05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t1</th>         <td>-0.0457</td>   <td>0.0028</td>   <td>-16.545</td> <td>0.0000</td>   <td>-0.0511</td>   <td>-0.0403</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t2</th>         <td>-0.0543</td>   <td>0.0031</td>   <td>-17.583</td> <td>0.0000</td>   <td>-0.0604</td>   <td>-0.0483</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t3</th>         <td>-0.0254</td>   <td>0.0045</td>   <td>-5.6924</td> <td>0.0000</td>   <td>-0.0342</td>   <td>-0.0167</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t4</th>         <td>-0.0022</td>   <td>0.0039</td>   <td>-0.5710</td> <td>0.5680</td>   <td>-0.0098</td>   <td>0.0054</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t5</th>         <td>-0.0095</td>   <td>0.0040</td>   <td>-2.3570</td> <td>0.0184</td>   <td>-0.0173</td>   <td>-0.0016</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>jail3</th>      <td>0.1493</td>    <td>0.0652</td>   <td>2.2893</td>  <td>0.0221</td>   <td>0.0215</td>    <td>0.2771</td>  \n",
+       "</tr>\n",
+       "</table><br/><br/>Endogenous: jail3<br/>Instruments: judge_pre_1, judge_pre_2, judge_pre_3, judge_pre_4, judge_pre_5, judge_pre_6, judge_pre_7<br/>Robust Covariance (Heteroskedastic)<br/>Debiased: False"
+      ],
+      "text/plain": [
+       "<class 'linearmodels.compat.statsmodels.Summary'>\n",
+       "\"\"\"\n",
+       "                          IV-2SLS Estimation Summary                          \n",
+       "==============================================================================\n",
+       "Dep. Variable:                  guilt   R-squared:                      0.4849\n",
+       "Estimator:                    IV-2SLS   Adj. R-squared:                 0.4849\n",
+       "No. Observations:              331971   F-statistic:                 3.199e+05\n",
+       "Date:                Sun, Mar 07 2021   P-value (F-stat)                0.0000\n",
+       "Time:                        13:31:29   Distribution:                  chi2(9)\n",
+       "Cov. Estimator:                robust                                         \n",
+       "                                                                              \n",
+       "                             Parameter Estimates                              \n",
+       "==============================================================================\n",
+       "            Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
+       "------------------------------------------------------------------------------\n",
+       "day        -7.952e-05  2.506e-05    -3.1730     0.0015     -0.0001   -3.04e-05\n",
+       "day2          1.4e-07  5.806e-08     2.4111     0.0159   2.619e-08   2.538e-07\n",
+       "bailDate      3.2e-05  1.902e-06     16.829     0.0000   2.828e-05   3.573e-05\n",
+       "t1            -0.0457     0.0028    -16.545     0.0000     -0.0511     -0.0403\n",
+       "t2            -0.0543     0.0031    -17.583     0.0000     -0.0604     -0.0483\n",
+       "t3            -0.0254     0.0045    -5.6924     0.0000     -0.0342     -0.0167\n",
+       "t4            -0.0022     0.0039    -0.5710     0.5680     -0.0098      0.0054\n",
+       "t5            -0.0095     0.0040    -2.3570     0.0184     -0.0173     -0.0016\n",
+       "jail3          0.1493     0.0652     2.2893     0.0221      0.0215      0.2771\n",
+       "==============================================================================\n",
+       "\n",
+       "Endogenous: jail3\n",
+       "Instruments: judge_pre_1, judge_pre_2, judge_pre_3, judge_pre_4, judge_pre_5, judge_pre_6, judge_pre_7\n",
+       "Robust Covariance (Heteroskedastic)\n",
+       "Debiased: False\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#--- Instrumental Variables Estimations\n",
+    "#-- 2sls main results\n",
+    "#- Min and Max Control formulas\n",
+    "min_formula = \"guilt ~ {control2} + [jail3 ~ {judge_pre}]\".format(control2=control2, judge_pre=judge_pre)\n",
+    "max_formula = \"\"\"guilt ~ {demo} + possess + {prior} + robbery + {off} + DUI1st + {control2} + drugSell + aggAss +\n",
+    "                    [jail3 ~ {judge_pre}]\"\"\".format(demo=demo,\n",
+    "                                                    prior=prior,\n",
+    "                                                    off=off,\n",
+    "                                                    control2=control2,\n",
+    "                                                   judge_pre=judge_pre)\n",
+    "min_iv = IV2SLS.from_formula(min_formula, data = judge).fit()\n",
+    "max_iv = IV2SLS.from_formula(max_formula, data = judge).fit()\n",
+    "\n",
+    "\n",
+    "print(\"IV\")\n",
+    "min_iv.summary"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>IV-2SLS Estimation Summary</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>          <td>guilt</td>      <th>  R-squared:         </th>  <td>0.5428</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Estimator:</th>             <td>IV-2SLS</td>     <th>  Adj. R-squared:    </th>  <td>0.5428</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>331971</td>      <th>  F-statistic:       </th> <td>4.057e+05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, Mar 07 2021</td> <th>  P-value (F-stat)   </th>  <td>0.0000</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:31:32</td>     <th>  Distribution:      </th> <td>chi2(34)</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Cov. Estimator:</th>        <td>robust</td>      <th>                     </th>     <td></td>     \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th></th>                          <td></td>         <th>                     </th>     <td></td>     \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<caption>Parameter Estimates</caption>\n",
+       "<tr>\n",
+       "        <td></td>         <th>Parameter</th> <th>Std. Err.</th> <th>T-stat</th>  <th>P-value</th> <th>Lower CI</th>  <th>Upper CI</th>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black</th>           <td>0.0592</td>    <td>0.0054</td>   <td>10.967</td>  <td>0.0000</td>   <td>0.0486</td>    <td>0.0697</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>age</th>             <td>0.0015</td>    <td>0.0001</td>   <td>13.463</td>  <td>0.0000</td>   <td>0.0013</td>    <td>0.0017</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>male</th>            <td>-0.0515</td>   <td>0.0070</td>   <td>-7.3976</td> <td>0.0000</td>   <td>-0.0652</td>   <td>-0.0379</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>white</th>           <td>0.1016</td>    <td>0.0036</td>   <td>28.099</td>  <td>0.0000</td>   <td>0.0946</td>    <td>0.1087</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>possess</th>         <td>-0.0568</td>   <td>0.0039</td>   <td>-14.739</td> <td>0.0000</td>   <td>-0.0643</td>   <td>-0.0492</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>priorCases</th>      <td>-0.0058</td>   <td>0.0003</td>   <td>-22.891</td> <td>0.0000</td>   <td>-0.0063</td>   <td>-0.0053</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>priorWI5</th>        <td>0.0267</td>    <td>0.0054</td>   <td>4.9923</td>  <td>0.0000</td>   <td>0.0162</td>    <td>0.0372</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>prior_felChar</th>   <td>-0.0073</td>   <td>0.0008</td>   <td>-9.4857</td> <td>0.0000</td>   <td>-0.0088</td>   <td>-0.0058</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>prior_guilt</th>     <td>0.0237</td>    <td>0.0010</td>   <td>22.757</td>  <td>0.0000</td>   <td>0.0217</td>    <td>0.0258</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>onePrior</th>        <td>0.0478</td>    <td>0.0035</td>   <td>13.613</td>  <td>0.0000</td>   <td>0.0409</td>    <td>0.0546</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>threePriors</th>     <td>-0.0002</td>   <td>0.0044</td>   <td>-0.0558</td> <td>0.9555</td>   <td>-0.0088</td>   <td>0.0083</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>robbery</th>         <td>-0.0971</td>   <td>0.0080</td>   <td>-12.175</td> <td>0.0000</td>   <td>-0.1127</td>   <td>-0.0815</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>fel</th>             <td>-0.0256</td>   <td>0.0098</td>   <td>-2.6104</td> <td>0.0090</td>   <td>-0.0448</td>   <td>-0.0064</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>mis</th>             <td>0.1331</td>    <td>0.0100</td>   <td>13.269</td>  <td>0.0000</td>   <td>0.1135</td>    <td>0.1528</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>sum</th>             <td>0.0656</td>    <td>0.0037</td>   <td>17.974</td>  <td>0.0000</td>   <td>0.0585</td>    <td>0.0728</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>F1</th>              <td>-0.0068</td>   <td>0.0104</td>   <td>-0.6559</td> <td>0.5119</td>   <td>-0.0272</td>   <td>0.0136</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>F2</th>              <td>0.0304</td>    <td>0.0097</td>   <td>3.1197</td>  <td>0.0018</td>   <td>0.0113</td>    <td>0.0495</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>F3</th>              <td>0.0944</td>    <td>0.0031</td>   <td>30.554</td>  <td>0.0000</td>   <td>0.0883</td>    <td>0.1004</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>M1</th>              <td>0.0097</td>    <td>0.0069</td>   <td>1.4168</td>  <td>0.1565</td>   <td>-0.0037</td>   <td>0.0231</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>M2</th>              <td>-0.0750</td>   <td>0.0040</td>   <td>-18.628</td> <td>0.0000</td>   <td>-0.0829</td>   <td>-0.0671</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>M3</th>              <td>0.1252</td>    <td>0.0047</td>   <td>26.564</td>  <td>0.0000</td>   <td>0.1159</td>    <td>0.1344</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>M</th>               <td>0.2607</td>    <td>0.0044</td>   <td>59.621</td>  <td>0.0000</td>   <td>0.2522</td>    <td>0.2693</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>DUI1st</th>          <td>0.0584</td>    <td>0.0061</td>   <td>9.6164</td>  <td>0.0000</td>   <td>0.0465</td>    <td>0.0703</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>day</th>           <td>-8.052e-05</td> <td>2.373e-05</td> <td>-3.3925</td> <td>0.0007</td>   <td>-0.0001</td>  <td>-3.4e-05</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>day2</th>           <td>2.427e-07</td> <td>5.473e-08</td> <td>4.4349</td>  <td>0.0000</td>  <td>1.355e-07</td> <td> 3.5e-07</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>bailDate</th>       <td>7.733e-06</td> <td>8.764e-07</td> <td>8.8233</td>  <td>0.0000</td>  <td>6.015e-06</td> <td>9.45e-06</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t1</th>              <td>-0.0841</td>   <td>0.0027</td>   <td>-30.731</td> <td>0.0000</td>   <td>-0.0895</td>   <td>-0.0788</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t2</th>              <td>-0.0753</td>   <td>0.0033</td>   <td>-22.738</td> <td>0.0000</td>   <td>-0.0818</td>   <td>-0.0688</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t3</th>              <td>-0.0392</td>   <td>0.0041</td>   <td>-9.5023</td> <td>0.0000</td>   <td>-0.0473</td>   <td>-0.0311</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t4</th>              <td>-0.0118</td>   <td>0.0036</td>   <td>-3.2958</td> <td>0.0010</td>   <td>-0.0188</td>   <td>-0.0048</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>t5</th>              <td>-0.0048</td>   <td>0.0040</td>   <td>-1.2026</td> <td>0.2291</td>   <td>-0.0127</td>   <td>0.0030</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>drugSell</th>        <td>0.0381</td>    <td>0.0061</td>   <td>6.2923</td>  <td>0.0000</td>   <td>0.0262</td>    <td>0.0500</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>aggAss</th>          <td>0.0085</td>    <td>0.0041</td>   <td>2.0867</td>  <td>0.0369</td>   <td>0.0005</td>    <td>0.0164</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>jail3</th>           <td>0.1813</td>    <td>0.0643</td>   <td>2.8181</td>  <td>0.0048</td>   <td>0.0552</td>    <td>0.3074</td> \n",
+       "</tr>\n",
+       "</table><br/><br/>Endogenous: jail3<br/>Instruments: judge_pre_1, judge_pre_2, judge_pre_3, judge_pre_4, judge_pre_5, judge_pre_6, judge_pre_7<br/>Robust Covariance (Heteroskedastic)<br/>Debiased: False"
+      ],
+      "text/plain": [
+       "<class 'linearmodels.compat.statsmodels.Summary'>\n",
+       "\"\"\"\n",
+       "                          IV-2SLS Estimation Summary                          \n",
+       "==============================================================================\n",
+       "Dep. Variable:                  guilt   R-squared:                      0.5428\n",
+       "Estimator:                    IV-2SLS   Adj. R-squared:                 0.5428\n",
+       "No. Observations:              331971   F-statistic:                 4.057e+05\n",
+       "Date:                Sun, Mar 07 2021   P-value (F-stat)                0.0000\n",
+       "Time:                        13:31:32   Distribution:                 chi2(34)\n",
+       "Cov. Estimator:                robust                                         \n",
+       "                                                                              \n",
+       "                               Parameter Estimates                               \n",
+       "=================================================================================\n",
+       "               Parameter  Std. Err.     T-stat    P-value    Lower CI    Upper CI\n",
+       "---------------------------------------------------------------------------------\n",
+       "black             0.0592     0.0054     10.967     0.0000      0.0486      0.0697\n",
+       "age               0.0015     0.0001     13.463     0.0000      0.0013      0.0017\n",
+       "male             -0.0515     0.0070    -7.3976     0.0000     -0.0652     -0.0379\n",
+       "white             0.1016     0.0036     28.099     0.0000      0.0946      0.1087\n",
+       "possess          -0.0568     0.0039    -14.739     0.0000     -0.0643     -0.0492\n",
+       "priorCases       -0.0058     0.0003    -22.891     0.0000     -0.0063     -0.0053\n",
+       "priorWI5          0.0267     0.0054     4.9923     0.0000      0.0162      0.0372\n",
+       "prior_felChar    -0.0073     0.0008    -9.4857     0.0000     -0.0088     -0.0058\n",
+       "prior_guilt       0.0237     0.0010     22.757     0.0000      0.0217      0.0258\n",
+       "onePrior          0.0478     0.0035     13.613     0.0000      0.0409      0.0546\n",
+       "threePriors      -0.0002     0.0044    -0.0558     0.9555     -0.0088      0.0083\n",
+       "robbery          -0.0971     0.0080    -12.175     0.0000     -0.1127     -0.0815\n",
+       "fel              -0.0256     0.0098    -2.6104     0.0090     -0.0448     -0.0064\n",
+       "mis               0.1331     0.0100     13.269     0.0000      0.1135      0.1528\n",
+       "sum               0.0656     0.0037     17.974     0.0000      0.0585      0.0728\n",
+       "F1               -0.0068     0.0104    -0.6559     0.5119     -0.0272      0.0136\n",
+       "F2                0.0304     0.0097     3.1197     0.0018      0.0113      0.0495\n",
+       "F3                0.0944     0.0031     30.554     0.0000      0.0883      0.1004\n",
+       "M1                0.0097     0.0069     1.4168     0.1565     -0.0037      0.0231\n",
+       "M2               -0.0750     0.0040    -18.628     0.0000     -0.0829     -0.0671\n",
+       "M3                0.1252     0.0047     26.564     0.0000      0.1159      0.1344\n",
+       "M                 0.2607     0.0044     59.621     0.0000      0.2522      0.2693\n",
+       "DUI1st            0.0584     0.0061     9.6164     0.0000      0.0465      0.0703\n",
+       "day           -8.052e-05  2.373e-05    -3.3925     0.0007     -0.0001    -3.4e-05\n",
+       "day2           2.427e-07  5.473e-08     4.4349     0.0000   1.355e-07     3.5e-07\n",
+       "bailDate       7.733e-06  8.764e-07     8.8233     0.0000   6.015e-06    9.45e-06\n",
+       "t1               -0.0841     0.0027    -30.731     0.0000     -0.0895     -0.0788\n",
+       "t2               -0.0753     0.0033    -22.738     0.0000     -0.0818     -0.0688\n",
+       "t3               -0.0392     0.0041    -9.5023     0.0000     -0.0473     -0.0311\n",
+       "t4               -0.0118     0.0036    -3.2958     0.0010     -0.0188     -0.0048\n",
+       "t5               -0.0048     0.0040    -1.2026     0.2291     -0.0127      0.0030\n",
+       "drugSell          0.0381     0.0061     6.2923     0.0000      0.0262      0.0500\n",
+       "aggAss            0.0085     0.0041     2.0867     0.0369      0.0005      0.0164\n",
+       "jail3             0.1813     0.0643     2.8181     0.0048      0.0552      0.3074\n",
+       "=================================================================================\n",
+       "\n",
+       "Endogenous: jail3\n",
+       "Instruments: judge_pre_1, judge_pre_2, judge_pre_3, judge_pre_4, judge_pre_5, judge_pre_6, judge_pre_7\n",
+       "Robust Covariance (Heteroskedastic)\n",
+       "Debiased: False\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "max_iv.summary"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from rpy2 import robjects\n",
+    "from rpy2.robjects import pandas2ri\n",
+    "from rpy2.robjects.packages import importr\n",
+    "pandas2ri.activate()\n",
+    "SteinIV = importr('SteinIV')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "<ipython-input-10-7d8aa8c8216d>:5: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "<ipython-input-10-7d8aa8c8216d>:8: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
+     ]
+    }
+   ],
+   "source": [
+    "#-- JIVE main results\n",
+    "#- minimum controls\n",
+    "y = judge['guilt']\n",
+    "X_min = judge[['jail3', 'day', 'day2', 't1', 't2', 't3', 't4', 't5', 'bailDate']]\n",
+    "X_min['intercept'] = 1\n",
+    "\n",
+    "Z_min = judge[judge_pre.split('+') + ['day', 'day2', 't1', 't2', 't3', 't4', 't5', 'bailDate']]\n",
+    "Z_min['intercept'] = 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "JIVE\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "\n",
+       "        <span>ListVector with 1 elements.</span>\n",
+       "        <table>\n",
+       "        <tbody>\n",
+       "        \n",
+       "          <tr>\n",
+       "            <th>\n",
+       "            est\n",
+       "            </th>\n",
+       "            <td>\n",
+       "            <rpy2.rinterface.FloatSexpVector object at 0x7ff77cd485c0> [RTYPES.REALSXP]\n",
+       "            </td>\n",
+       "          </tr>\n",
+       "        \n",
+       "        </tbody>\n",
+       "        </table>\n",
+       "        "
+      ],
+      "text/plain": [
+       "<rpy2.robjects.vectors.ListVector object at 0x7ff782e35900> [RTYPES.VECSXP]\n",
+       "R classes: ('list',)\n",
+       "[FloatSexpVector]\n",
+       "  est: <class 'rpy2.rinterface.FloatSexpVector'>\n",
+       "  <rpy2.rinterface.FloatSexpVector object at 0x7ff77cd81ec0> [RTYPES.REALSXP]"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "y = robjects.globalenv['y'] = y\n",
+    "X_min = robjects.globalenv['X_min'] = np.array(X_min)\n",
+    "Z_min = robjects.globalenv['Z_min'] = np.array(Z_min)\n",
+    "\n",
+    "print(\"JIVE\")\n",
+    "SteinIV.jive_est(y = y, X=X_min, Z=Z_min)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "<ipython-input-12-f819811e0eab>:10: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "<ipython-input-12-f819811e0eab>:21: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
+     ]
+    }
+   ],
+   "source": [
+    "X_max = judge[['jail3', 'white', 'age', 'male', 'black',\n",
+    "         'possess', 'robbery', \n",
+    "         'prior_guilt', 'onePrior', 'priorWI5', 'prior_felChar', 'priorCases',\n",
+    "         'DUI1st', 'drugSell', 'aggAss', 'fel', 'mis', 'sum',\n",
+    "         'threePriors',\n",
+    "         'F1', 'F2', 'F3',\n",
+    "         'M', 'M1', 'M2', 'M3',\n",
+    "         'day', 'day2', 'bailDate', \n",
+    "         't1', 't2', 't3', 't4', 't5']]\n",
+    "X_max['intercept'] = 1\n",
+    "\n",
+    "Z_max = judge[judge_pre.split('+') + ['white', 'age', 'male', 'black',\n",
+    "         'possess', 'robbery', \n",
+    "         'prior_guilt', 'onePrior', 'priorWI5', 'prior_felChar', 'priorCases',\n",
+    "         'DUI1st', 'drugSell', 'aggAss', 'fel', 'mis', 'sum',\n",
+    "         'threePriors',\n",
+    "         'F1', 'F2', 'F3',\n",
+    "         'M', 'M1', 'M2', 'M3',\n",
+    "         'day', 'day2', 'bailDate', \n",
+    "         't1', 't2', 't3', 't4', 't5']]\n",
+    "Z_max['intercept'] = 1\n",
+    "X_max = robjects.globalenv['X_max'] = np.array(X_max)\n",
+    "Z_max = robjects.globalenv['Z_max'] = np.array(Z_max)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "\n",
+       "        <span>ListVector with 1 elements.</span>\n",
+       "        <table>\n",
+       "        <tbody>\n",
+       "        \n",
+       "          <tr>\n",
+       "            <th>\n",
+       "            est\n",
+       "            </th>\n",
+       "            <td>\n",
+       "            <rpy2.rinterface.FloatSexpVector object at 0x7ff77b1c8c80> [RTYPES.REALSXP]\n",
+       "            </td>\n",
+       "          </tr>\n",
+       "        \n",
+       "        </tbody>\n",
+       "        </table>\n",
+       "        "
+      ],
+      "text/plain": [
+       "<rpy2.robjects.vectors.ListVector object at 0x7ff77b4daf80> [RTYPES.VECSXP]\n",
+       "R classes: ('list',)\n",
+       "[FloatSexpVector]\n",
+       "  est: <class 'rpy2.rinterface.FloatSexpVector'>\n",
+       "  <rpy2.rinterface.FloatSexpVector object at 0x7ff77b1c8d00> [RTYPES.REALSXP]"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "SteinIV.jive_est(y = y, X = X_max, Z = Z_max)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTION\n",
+    "- Interpret the coefficient on our two IV estimators?  How do they compare to our OLS estimate?\n",
+    "- What is your conclusion about the effect that cash bail has on adjudication?  Speculate about the channels by which cash bail has this effect. \n",
+    "- Describe the four sub-populations (e.g., always takers, never takers, defiers and compliers) in the context of Stevenson's study.\n",
+    "- Discuss the plausibility of each of the 5 IV assumptions in Stevenson's case.  \n",
+    "- Draw a DAG that must be true for Stevenson's JIVE estimates to be consistent?  Which assumptions are contained in this DAG and which ones are not easily visualized? \n",
+    "- Assume judge A is stricter than judge B.  Monotonicity requires that if judge B sets a lower bail amount for that individual, then judge A will always set a higher for that individual hypothetically than judge B.  Provide some examples where you think this may be violated.  \n",
+    "\n",
+    "\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  },
+  "toc": {
+   "base_numbering": 1,
+   "nav_menu": {},
+   "number_sections": true,
+   "sideBar": true,
+   "skip_h1_title": false,
+   "title_cell": "Table of Contents",
+   "title_sidebar": "Contents",
+   "toc_cell": false,
+   "toc_position": {},
+   "toc_section_display": true,
+   "toc_window_display": false
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/Python/Matching_and_Subclassification.ipynb b/Python/Matching_and_Subclassification.ipynb
new file mode 100644
index 0000000..92f6f7f
--- /dev/null
+++ b/Python/Matching_and_Subclassification.ipynb
@@ -0,0 +1,557 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Matching and Subclassification** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import plotnine as p\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def read_data(file):\n",
+    "    full_path = \"https://raw.github.com/scunning1975/mixtape/master/\" + file\n",
+    "    \n",
+    "    return pd.read_stata(full_path)\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## Simple Difference in Outcomes\n",
+    "titanic = read_data(\"titanic.dta\")\n",
+    "\n",
+    "titanic['d'] = 0\n",
+    "titanic.loc[titanic['class']=='1st class', 'd'] = 1\n",
+    "\n",
+    "titanic['sex_d'] = 0\n",
+    "titanic.loc[titanic['sex']=='man', 'sex_d'] = 1\n",
+    "\n",
+    "titanic['age_d'] = 0\n",
+    "titanic.loc[titanic['age']=='adults', 'age_d'] = 1\n",
+    "\n",
+    "titanic['survived_d'] = 0\n",
+    "titanic.loc[titanic['survived']=='yes', 'survived_d'] = 1\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The simple difference in outcomes is 35.38%\n"
+     ]
+    }
+   ],
+   "source": [
+    "ey0 = titanic.loc[titanic['d']==0, 'survived_d'].mean()\n",
+    "ey1 = titanic.loc[titanic['d']==1, 'survived_d'].mean()\n",
+    "\n",
+    "sdo = ey1 - ey0\n",
+    "print(\"The simple difference in outcomes is {:.2%}\".format(sdo))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The weigthted average treatment effect estimate is 16.09%\n"
+     ]
+    }
+   ],
+   "source": [
+    "## Weighted Average Treatment Effect \n",
+    "titanic['s'] = 0 \n",
+    "titanic.loc[(titanic.sex_d == 0) & (titanic.age_d==1), 's'] = 1\n",
+    "titanic.loc[(titanic.sex_d == 0) & (titanic.age_d==0), 's'] = 2\n",
+    "titanic.loc[(titanic.sex_d == 1) & (titanic.age_d==1), 's'] = 3\n",
+    "titanic.loc[(titanic.sex_d == 1) & (titanic.age_d==0), 's'] = 4\n",
+    "\n",
+    "obs = titanic.shape[0]\n",
+    "\n",
+    "def weighted_avg_effect(df):\n",
+    "    diff = df[df.d==1].survived_d.mean() - df[df.d==0].survived_d.mean()\n",
+    "    weight = df[df.d==0].shape[0]/obs\n",
+    "    return diff*weight\n",
+    "\n",
+    "wate = titanic.groupby('s').apply(weighted_avg_effect).sum()\n",
+    "\n",
+    "print(\"The weigthted average treatment effect estimate is {:.2%}\".format(wate))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Questions\n",
+    "- Using the simple difference in outcomes, how much does the probability of survival increase for first-class passengers relative to some control group?\n",
+    "- Explain in your own words what stratifying on gender and age did for this difference in outcomes between treatment and control?\n",
+    "- After stratifying on gender and age, what happens to the difference in probability of survival between first-class and non-first-class passengers?\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Training Example\n",
+    "\n",
+    "First, we will look at the distribution of age between the treated and non-treated groups"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/layer.py:372: PlotnineWarning: stat_bin : Removed 14 rows containing non-finite values.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAGvCAYAAAC9yRSTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAhqklEQVR4nO3de3BU9fnH8c/JJrtJNgkQ5RIMhItIuQha68hVIKJtQeOlmo5AaVSUglPGWkHllyGJCra1VB2u6XQGi4pi0WEs1UoFpRqoRR0UaOR+EdmCsUFiSNjN5vz+QBbXBITNbs7ul/frL3f3cPbhzMP23d1cLNu2bQEAABgqyekBAAAAYonYAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABgt2ekBWkNVVVXMzu12u+X3+2N2/kRiWZbS0tJUV1cnfjA3u/FN7EY4duMUdqMp9uMEy7J0wQUXROVcvLPTQh6Px+kR4kZSUpLS09OVlMRaSezGN7Eb4diNU9iNptiPE6K5E2wXAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMFrC/bqIQCCgxYsX66OPPlJNTY0uvPBC3XbbbRo5cqTTowEAgDiUcLETDAaVnZ2txx57TB06dNAnn3yiRx55RJ06ddL3vvc9p8cDAABxJuE+xkpNTdX48ePVqVMnJSUlqW/fvurTp48qKyudHg0AAMShhIudb6uvr9fOnTuVl5fn9CgAACAOJXTs2Latp59+Wr169dLll1/u9DgAACAOJdzX7Jxk27YWLlyoL774Qo888ogsywo95vP55PP5Qrc9Ho86d+4ckzksy5LL5YrJuRPNyevA9TiB3TiF3QjHbpzCbjTFfpwQzWtg2bZtR+1srcS2bS1evFg7d+7Uo48+qvT09LDHS0tLVVZWFro9c+ZMzZ49OyazjBs3LibnjaVly5Y5PcI54zoDACKVkO/slJeXa9u2bXrssceahI4kTZ48WQUFBaHbHo9H1dXVrTliXIvVtXC5XMrKytLRo0cVDAZj8hyJxO/3q7a21ukx4gK7Ec7r9bIbX2M3mmI/Tji5G9GQcLFz+PBhvfbaa0pJSdGdd94Zuv/WW29VYWGhJCknJ0c5OTmhx6qqqvhH9A2xvhbBYJDrrRPvQHIdwrEbJ7AbTbEbp7Af0ZdwsdOhQwe9+uqrTo8BAAASREJ/NxYAAMB3IXYAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRkp0eoDW43W55PB6nx4gbmZmZMTmvZVmSJK/XK9u2Y/IciSQ5OTlm1zrRsBvh2I1T2I2m2I8TTu5GNJwXseP3++X3+50eI27U1NTE5Lwul0tut1u1tbUKBoMxeY5E0tDQELNrnWjYjXCZmZnsxtfYjabYjxNO7kY08DEWAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjJbs9ACrVq3S2rVrtXfvXg0ePFjTp08/7bEFBQXyeDyyLEuS1LdvX5WWlrbSpAAAIBE5HjvZ2dkqLCzUpk2bVFNT853HP/nkk8rNzW2FyQAAgAkcj50hQ4ZIknbv3n1WsQMAAHAuHI+dc1VcXKxgMKhevXqpqKhIXbt2dXokAAAQxxIqdubMmaPevXsrEAjolVde0axZs7Rw4UKlp6c7PRoAAIhTCRU7/fv3lySlpKRowoQJeuutt1RZWakrrrgi7Difzyefzxe67fF41Llz51adNZ65XK6YnjdW5080lmVxLb7GboRjN05hN5piP06I5jVIqNj5tpPflfVt5eXlKisrC92eOXOmZs+e3Vpjxb127drF9PxZWVkxPX+icLvdcrvdTo8RV9iNU9iNcOxGOPYjuhyPnWAwqGAwqMbGRjU2Nsrv9yspKUnJyeGj7d+/X4FAQN26dVNDQ4Nefvll+f1+9e7du8k5J0+erIKCgtBtj8ej6urqmP9dEkWsroXL5VJWVpaOHj2qYDAYk+dIJH6/X7W1tU6PERfYjXBer5fd+Bq70RT7ccLJ3YgGx2Nn+fLlevHFF0O3KyoqlJ+fr/vuu0+FhYUqKSlRv379dOTIES1atEhVVVVyu926+OKLVVZWpoyMjCbnzMnJUU5OTuh2VVUV/4i+IdbX4mTAnu9s2+Y6fAu7cQK70RS7cQr7EX2Ox864ceM0bty4Zh976aWXQv89YMAALVq0qLXGAgAAhuDXRQAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjJTg/QGtxutzwej9NjxI3MzMyYnNeyLEmS1+uVbdsxeY5EkpycHLNrnWjYjXDsxinsRlPsxwkndyMazovY8fv98vv9To8RN2pqamJyXpfLJbfbrdraWgWDwZg8RyJpaGiI2bVONOxGuMzMTHbja+xGU+zHCSd3Ixr4GAsAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARosodvLz8/XJJ580+9j27duVn5/foqEAAACiJaLYefvtt3X06NFmHzt69Kj++c9/tmgoAACAaIn4YyzLspq9f/369erQoUPEAwEAAERT8tke+Pjjj+vxxx+XdCJ0Ro0apaSk8FY6fvy4GhoaNHXq1OhOCQAAEKGzjp0hQ4bo17/+tWzb1iOPPKLbb79dubm5Yce43W716dNHN9xwQ9QHBQAAiMRZx86IESM0YsQISSfe2bn77rvVuXPnmA0GAAAQDWcdO99UUlIS7TkAAABiIqLYaWxs1J/+9CetWLFCBw4cUH19fdjjlmVp165dURkQAACgJSKKnQcffFBz587V0KFDNXz4cLnd7mjPBQAAEBURxc7zzz+v0tJSzZo1K9rzAAAARFVEP2envr5eQ4cOjfYsAAAAURdR7IwfP15//etfoz0LAABA1EX0MdagQYNUXFysQ4cO6dprr1Xbtm2bHHPLLbe0dDYAAIAWiyh2fvazn0mS9u3bp+XLlzd53LIsBYPBlk0GAAAQBRHFzp49e6I9BwAAQExEFDt5eXnRngMAACAmIoqd/fv3f+cxXbt2jeTU32nVqlVau3at9u7dq8GDB2v69OkxeR4AAGCGiGKnW7dusizrjMfE6mt2srOzVVhYqE2bNqmmpiYmzwEAAMwRUez85S9/aXLf//73P61evVobN27U7NmzWzzY6QwZMkSStHv3bmIHAAB8p4hi5yc/+Umz999999361a9+pYqKCo0fP75FgwEAAERDRD9U8EzGjh2rF198MdqnBQAAiEhE7+ycyfr165Wamhrt054Tn88nn88Xuu3xeNS5c2cHJ4ovLpcrpueN1fkTjWVZXIuvsRvh2I1T2I2m2I8TonkNIoqdadOmNbnP7/ersrJS7777rh544IEWD9YS5eXlKisrC92eOXNmTL+OKNHce++9To9wXigqKnJ6BOC8t2zZMqdHOGfjxo1zeoRzFu/XOaLYae73YqWmpio3N1cLFy7UpEmTWjzY6QSDQQWDQTU2NqqxsVF+v19JSUlKTj71V5k8ebIKCgpCtz0ej6qrq2M2EwAgPvHa3zpicZ1dLpeysrKicq6E+wnKy5cvD/uaoIqKCuXn5+u+++4L3ZeTk6OcnJzQ7aqqKn59BQCch3jtbx3xfp1b/DU7tm3rq6++UkZGxnf+7J1oGDduXEK+xQcAAJwR8XdjrVu3Tvn5+UpLS1Pbtm2Vlpama665Ru+880405wMAAGiRiN7Z+cc//qExY8bokksu0cMPP6xOnTrJ5/NpxYoVuuaaa/Taa69p9OjR0Z4VAADgnEUUO8XFxRozZoxWrlwZ9tFVSUmJbrrpJhUXFxM7AAAgLkT0MdbmzZs1ZcqUJl+jY1mWpkyZoo8//jgqwwEAALRURLGTkZGhzz77rNnHDhw4oIyMjBYNBQAAEC0RxU5BQYEeeughvfHGG2H3r169Wv/3f/+nG2+8MSrDAQAAtFREX7PzxBNPaPPmzfrxj3+srKwsdezYUYcOHVJNTY2uvPJKPfHEE9GeEwAAICIRxU67du20YcMGrVq1Su+++66qq6uVnZ2tYcOGaezYsUpKivrvFwUAAIhIRLGzZs0a7d+/X3fccUfYr2WQpGeeeUZ5eXkaNWpUVAYEAABoiYjegikuLtahQ4eafezzzz9XcXFxi4YCAACIlohiZ+vWrfrBD37Q7GPf//73tXXr1hYNBQAAEC0RxY5lWfryyy+bfay6ujrufyEYAAA4f0QUO1dddZUWLFgg27bD7rdtWwsXLtRVV10VleEAAABaKqIvUC4rK9OoUaM0YMAAFRUVKScnRwcPHtTSpUu1fft2vf3221EeEwAAIDIRxc7gwYO1Zs0azZgxQw8++KAaGxuVlJQUun/QoEHRnhMAACAiEcWOJA0dOlQVFRWqq6tTdXW12rZtq/T09GjOBgAA0GIRx85JaWlpSktLi8YsAAAAUcePOgYAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGS3Z6gNbgdrvl8XicHgMA0MoyMzOdHuG8EIvrbFlW1M51XsSO3++X3+93egwAQCurqalxeoTzQiyus8vlktvtjsq5+BgLAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARkt2egBJ+uqrr7RgwQJ9+OGHSktLU2FhocaMGdPssQUFBfJ4PLIsS5LUt29flZaWtuK0AAAgkcRF7JSXlysYDGrJkiXy+XyaNWuWcnNzNWDAgGaPf/LJJ5Wbm9vKUwIAgETk+MdY9fX1qqio0IQJE5Senq6ePXsqPz9fb775ptOjAQAAAzj+zs5nn30mSeratWvovh49emjlypWn/TPFxcUKBoPq1auXioqKwv4sAADANzkeO/X19UpLSwu7z+v1qq6urtnj58yZo969eysQCOiVV17RrFmztHDhQqWnp7fGuAAAIME4HjupqalNwqa2trZJAJ3Uv39/SVJKSoomTJigt956S5WVlbriiitCx/h8Pvl8vtBtj8ejzp07x2B6AEA8c7lcTo9wXojFdY7mOR2PnYsuukiS9Omnn6pLly6SpD179igvL++s/vzJ78r6pvLycpWVlYVuz5w5U7Nnz47CtACARNKuXTunRzgvxPt1djx2UlNTNXToUD3//POaNm2aDh06pDVr1mjGjBlNjt2/f78CgYC6deumhoYGvfzyy/L7/erdu3fYcZMnT1ZBQUHotsfjUXV1dcz/LgCA+MJrf+uIxXV2uVzKysqKyrkcjx3pRJzMnz9fRUVFSk9P1/jx4zVw4EBJUmFhoUpKStSvXz8dOXJEixYtUlVVldxuty6++GKVlZUpIyMj7Hw5OTnKyckJ3a6qqlIwGGzVvxMAwHm89reOeL/OcRE7GRkZeuihh5p97KWXXgr994ABA7Ro0aLWGgsAABjA8Z+zAwAAEEvEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMFqy0wO0BrfbLY/H4/QYAIBWlpmZ6fQI54VYXGfLsqJ2rvMidvx+v/x+v9NjAABaWU1NjdMjnBdicZ1dLpfcbndUzsXHWAAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADAasQMAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDAACMRuwAAACjETsAAMBoxA4AADBastMDnKuvvvpKCxYs0Icffqi0tDQVFhZqzJgxTo8FAADiVMLFTnl5uYLBoJYsWSKfz6dZs2YpNzdXAwYMcHo0AAAQhxLqY6z6+npVVFRowoQJSk9PV8+ePZWfn68333zT6dEAAECcSqjY+eyzzyRJXbt2Dd3Xo0cP7du3z6mRAABAnEuo2Kmvr1daWlrYfV6vV3V1dQ5NBAAA4l1Cfc1Oampqk7Cpra1tEkA+n08+ny902+PxqHPnzq0yIwAgfrhcLqdHOC/E4jpH85wJFTsXXXSRJOnTTz9Vly5dJEl79uxRXl5e2HHl5eUqKysL3Z45c6Zmz54dk5mWLVsWk/MCAM5P/O9K9Fm2bdtOD3Eu5s6dq0AgoGnTpunQoUMqLi7WjBkzNHDgwNAxrfnOjtfrVW1tbUzOnWhcLpeysrJ09OhRBYNBp8dxHLtxCrsRjt04hd1oiv044eRuRENCvbMjSZMnT9b8+fNVVFSk9PR0jR8/Pix0JCknJ0c5OTmh21VVVTH7R2TbNv9AvyUYDHJNxG40h904gd1oit04hf2IvoSLnYyMDD300ENOjwEAABJEQn03FgAAwLkidgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0YgdAABgNGIHAAAYjdgBAABGI3YAAIDRiB0AAGA0YgcAABjNsm3bdnoImMHn86m8vFyTJ08O+63zALuB02E3cDrR3A3e2UHU+Hw+lZWVyefzOT0K4gy7gdNhN3A60dwNYgcAABiN2AEAAEYjdhA1OTk5Kikp4XN3NMFu4HTYDZxONHeDL1AGAABG450dAABgNGIHAAAYLdnpAZB4AoGAFi9erI8++kg1NTW68MILddttt2nkyJGSpH379mnevHnau3evOnXqpClTpqhfv37ODo1W8V27MWnSJB05ckRJSSf+f1b79u21YMECBydGa5o/f77ef/991dXVKTMzU9ddd50KCwsl8bqBM+9Hi187bOAc1dXV2c8995zt8/nsYDBob9261f7pT39qV1ZW2oFAwJ40aZK9YsUK2+/322vXrrVvv/12u6amxumx0QrOtBu2bdt33XWX/f777zs8JZyyb98+u76+3rZt2z58+LA9depU+5133uF1A7Ztn34/bLvlrx18jIVzlpqaqvHjx6tTp05KSkpS37591adPH1VWVmrz5s06fvy4br75ZqWkpGjUqFHq2LGj1q9f7/TYaAVn2g2ga9eu8ng8oduWZengwYO8bkDS6fcjGvgYCy1WX1+vnTt36oYbbtD+/fuVl5cXeqtRkrp37679+/c7OCGc8s3dOOmpp56Sbdvq2rWrJkyYoL59+zo4IVrbn//8Z61atUrHjx9Xhw4dNGrUKK1fv57XDUhqfj9OaslrB7GDFrFtW08//bR69eqlyy+/XNu3b5fX6w07xuv16tixYw5NCKd8ezck6f7771fPnj0lSWvWrFFZWZnmzZunDh06ODkqWtHPf/5zTZw4UTt37tR7770nr9eruro6Xjcgqfn9kFr+2sHHWIiYbdtauHChvvjiC82YMUOWZSktLa3JC9SxY8eUlpbm0JRwQnO7IUl9+/aVx+ORx+PRmDFj1KNHD33wwQcOT4vWZlmWevXqpeTkZL3wwgu8biDMt/dDavlrB7GDiNi2rcWLF2v37t0qLS1VamqqpBOfue7bt0+NjY2hY/fs2aOuXbs6NSpa2el2ozlJSUmy+bmm563Gxkb5fD5eN9Csk/vRnHN97SB2EJHy8nJt27ZNZWVlSk9PD91/6aWXKiUlRStXrlQgENC6dev03//+V4MHD3ZwWrSm0+3G559/rq1btyoQCCgQCOiNN97Qjh07Qh9xwWy1tbV66623dOzYMTU2Nuo///mPXn/9dV122WW8buCM+xGN1w5+XQTO2eHDhzVp0iSlpKTI5XKF7r/11ltVWFiovXv3av78+dq7d686duyoKVOmqH///g5OjNZypt0YNGiQ5s6dK5/Pp+TkZHXp0kUTJkzQpZde6uDEaC3Hjh3TnDlztGvXLjU2Nio7O1ujR4/WLbfcIsuyeN04z51pPz799NMWv3YQOwAAwGh8jAUAAIxG7AAAAKMROwAAwGjEDgAAMBqxAwAAjEbsAAAAoxE7AADAaMQOAAAwGrEDIO5t2rRJpaWljvwW7Lfffltz5sxp9ecFED3EDoC4t2nTJpWVlRE7ACJC7AAwSl1dndMjAIgzxA6AZm3YsEEFBQXq3LmzvF6vLrvsMj377LNhx2zdulVXX321UlNT1bNnTy1dulTXX3+9Ro4cGXZcZWWlbrzxRrVp00Zer1djx47Vrl27zmqOZ555RnfccYckqX379rIsS926dQs9ZlmWNmzYoGuvvVZer1cPPPCAJOnAgQOaMGGCLrzwQqWlpenqq6/WBx98EHbupUuXatiwYcrOzla7du00cuRI/fvf/w49XlpaqrKyMtXW1sqyLFmW1eTvBiD+JTs9AID4tG/fPg0dOlS/+MUvlJqaqoqKCt11112ybVsTJ05UXV2drrvuOrVt21bPPfecJKmkpERHjhxRr169QufZvXu3hgwZov79++uZZ55RUlKSZs+erWuuuUbbtm2Tx+M54xxjx45VcXGxHnvsMf39739XmzZtmvyZ8ePH65577tHMmTOVlpam6upqDRs2TBkZGZo3b57atGmjefPmKT8/Xzt27FCHDh0kSXv37tXEiRPVs2dP+f1+LVu2TFdffbU+/vhjXXLJJZo0aZIOHDigZcuWae3atZKkrKysaF5mAK3BBoDv0NjYaAcCAfuee+6xBw8ebNu2bS9YsMBOSkqyd+/eHTpu586ddlJSkj1ixIjQfRMnTrS7d+9u19XVhe47fPiw7fV67QULFpzV8y9ZssSWZH/++efN3v+73/0u7P5Zs2bZbdq0sQ8dOhS6r76+3s7NzbWnT5/e7HMEg0E7EAjYvXv3th9++OHQ/SUlJbbX6z2rOQHEJz7GAtCs6upqTZs2TXl5eUpJSVFKSor++Mc/avv27ZKkjRs3asCAAerevXvoz/Ts2VP9+/cPO8/q1at14403Kjk5WQ0NDWpoaFC7du00cOBAbdy4MSqzjhkzpslzjho1StnZ2aHndLlcGj58eNhzVlZW6uabb1bHjh3lcrmUkpKibdu2hf6OAMzAx1gAmlVUVKT169dr1qxZ6tevn7KysrRo0SItX75ckuTz+dS+ffsmf65Dhw4KBAKh21VVVXrqqaf01FNPNTk2LS0tKrOe/Fjqm8/5r3/9SykpKU2O7dmzpySppqZG1113ndq3b68//OEPysvLU2pqqiZNmqT6+vqozAUgPhA7AJqor6/X3/72N82dO1e//OUvQ/c3NjaG/jsnJ0ebNm1q8mcPHz6sdu3ahW5nZ2dr7Nixmjp1apNjMzMzozKvZVlht7Ozs/WjH/1Ijz76aJNjT369z4YNG3TgwAGtWrVKAwcODD3+5ZdfKjc3NypzAYgPxA6AJo4fP65gMCi32x26r6amRq+++mro9pVXXqmlS5dqz549oY+ydu3apS1btmj48OGh40aPHq0tW7bo8ssvl8vlimiek3Oc7Tsuo0eP1nPPPac+ffrI6/U2e8zJb1H/5t9x/fr12rt3r/r16xf23MePH49obgDxga/ZAdBEmzZtdOWVV+o3v/mNVqxYoZUrV+raa69VmzZtQsfccccd6tSpk66//nq9/PLLWrFihQoKCtSpUyclJZ16aSkrK9OOHTv0wx/+UC+99JLWrVun5cuXa+rUqXrhhRfOap4+ffpIkhYsWKD33ntPmzdvPuPx999/vyzL0ogRI/Tss89q3bp1WrFihaZPn64nn3xSkjRo0CBlZGTo3nvv1erVq7VkyRLdfvvtuuiii5o8d0NDg55++mlt3LhR27ZtO6uZAcQRp79CGkB82rFjhz1q1Cg7PT3d7tKli/3EE080+c6kLVu22MOGDbPdbrfdvXt3e8mSJfbw4cPtm266Kexc27dvtwsLC+0LLrjA9ng8drdu3eyJEyfaW7ZsOet5SktL7dzcXDspKcnOy8uzbfv036Vl27bt8/nsu+66y87JybHdbredm5tr33rrrXZFRUXomNdff93u16+fnZqaag8YMMB+7bXX7BEjRthjx44NHRMIBOypU6faHTt2tC3LCvtOMwCJwbJt23Y6uACY4YsvvlCPHj10//33q6SkxOlxAEASX7MDoAV++9vfqmPHjurWrZt8Pp9+//vfq7GxUXfeeafTowFACLEDIGIul0uzZ8/WgQMHlJycrKuuukpr165Vly5dzvocjY2NYd/l1dxzfPu7rQDgXPAxFgBHnfz9U6ezZMkSFRUVtd5AAIxD7ABw1MGDB3Xw4MHTPt69e3ddcMEFrTgRANMQOwAAwGj8nB0AAGA0YgcAABiN2AEAAEYjdgAAgNGIHQAAYDRiBwAAGI3YAQAARiN2AACA0f4f/JzeO/hIHsIAAAAASUVORK5CYII=\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8765026733050)>"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "training_example = read_data(\"training_example.dta\") \n",
+    "\n",
+    "p.ggplot(training_example, p.aes(x='age_treat')) +\\\n",
+    "  p.stat_bin(bins = 10)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/layer.py:372: PlotnineWarning: stat_bin : Removed 4 rows containing non-finite values.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8765026732969)>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(training_example, p.aes(x='age_control')) +\\\n",
+    "  p.geom_histogram(bins = 10)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "- Compare the distribution of ages between the treated and the control groups. How do they differ, if at all?\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "training_bias_reduction = read_data(\"training_bias_reduction.dta\") \n",
+    "\n",
+    "training_bias_reduction['Y1'] = 0\n",
+    "training_bias_reduction.loc[training_bias_reduction['Unit'].isin(range(1,5)), 'Y1'] = 1\n",
+    "training_bias_reduction['Y0'] = (4,0,5,1,4,0,5,1)\n",
+    "\n",
+    "\n",
+    "train_reg = sm.OLS.from_formula('Y ~ X', training_bias_reduction).fit()\n",
+    "training_bias_reduction['u_hat0'] = train_reg.predict(training_bias_reduction)\n",
+    "training_bias_reduction = training_bias_reduction[['Unit', 'Y1', 'Y0', 'Y', 'D', 'X', 'u_hat0']]\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## National Supported Work Demonstration Experiment\n",
+    "\n",
+    "To compare results, let's first look at the treatment effect identified by a true experiment.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The experimental ATE estimate is 1794.35\n"
+     ]
+    }
+   ],
+   "source": [
+    "nsw_dw = read_data('nsw_mixtape.dta')\n",
+    "\n",
+    "mean1 = nsw_dw[nsw_dw.treat==1].re78.mean()\n",
+    "mean0 = nsw_dw[nsw_dw.treat==0].re78.mean()\n",
+    "ate = np.unique(mean1 - mean0)[0]\n",
+    "print(\"The experimental ATE estimate is {:.2f}\".format(ate))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- How do you interpret the above estimated ATE?\n",
+    "- Say you were interested in the ATT.  Can you report the ATT from a randomized experiment?  If so, what is it? If not, why not?\n",
+    "\n",
+    "\n",
+    "\n",
+    "Now, lets turn to a non-experimental control group. We first have to load the data from the CPS. and estimate the propensity score\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Prepare data for logit \n",
+    "nsw_dw_cpscontrol = read_data('cps_mixtape.dta')\n",
+    "\n",
+    "nsw_dw_cpscontrol = pd.concat((nsw_dw_cpscontrol, nsw_dw))\n",
+    "nsw_dw_cpscontrol[['u74', 'u75']] = 0\n",
+    "nsw_dw_cpscontrol.loc[nsw_dw_cpscontrol.re74==0, 'u74'] = 1\n",
+    "nsw_dw_cpscontrol.loc[nsw_dw_cpscontrol.re75==0, 'u75'] = 1\n",
+    "# estimating propensity score\n",
+    "logit_nsw = smf.glm(formula=\"\"\"treat ~ age + age**2 + age**3 + educ + educ**2 + \n",
+    "                    marr + nodegree + black + hisp + re74 + re75 + u74 + u75 + educ*re74\"\"\", \n",
+    "                    family=sm.families.Binomial(),\n",
+    "                   data=nsw_dw_cpscontrol).fit()\n",
+    "                  \n",
+    "nsw_dw_cpscontrol['pscore'] = logit_nsw.predict(nsw_dw_cpscontrol)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "treat\n",
+       "0.0    0.009212\n",
+       "1.0    0.190701\n",
+       "Name: pscore, dtype: float64"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "nsw_dw_cpscontrol.groupby('treat')['pscore'].mean()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/facets/facet.py:549: PlotnineWarning: If you need more space for the x-axis tick text use ... + theme(subplots_adjust={'wspace': 0.25}). Choose an appropriate value for 'wspace'.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8765026990550)>"
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# histogram\n",
+    "p.ggplot(nsw_dw_cpscontrol, p.aes(x='pscore')) +\\\n",
+    "    p.geom_histogram(bins=50) +\\\n",
+    "    p.facet_wrap(\"treat\", scales='free')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Compare the mean propensity score between the treated and the control groups. What does this reveal about the two groups?\n",
+    "- Compare the distribution of propensity scores between the treated and the control groups. How do they differ, if at all?\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Treatment Effect (non-normalized, all data): -11535.55\n"
+     ]
+    }
+   ],
+   "source": [
+    "# continuation\n",
+    "N = nsw_dw_cpscontrol.shape[0]\n",
+    "\n",
+    "# Manual with non-normalized weights using all data\n",
+    "nsw_dw_cpscontrol = nsw_dw_cpscontrol \n",
+    "nsw_dw_cpscontrol['d1'] = nsw_dw_cpscontrol.treat/nsw_dw_cpscontrol.pscore\n",
+    "nsw_dw_cpscontrol['d0'] = (1-nsw_dw_cpscontrol.treat)/(1-nsw_dw_cpscontrol.pscore)\n",
+    "\n",
+    "\n",
+    "s1 = nsw_dw_cpscontrol.d1.sum()\n",
+    "s0 = nsw_dw_cpscontrol.d0.sum()\n",
+    "\n",
+    "nsw_dw_cpscontrol['y1'] = nsw_dw_cpscontrol.treat * nsw_dw_cpscontrol.re78 / nsw_dw_cpscontrol.pscore\n",
+    "nsw_dw_cpscontrol['y0'] = (1 - nsw_dw_cpscontrol.treat) * nsw_dw_cpscontrol.re78 / (1 - nsw_dw_cpscontrol.pscore)\n",
+    "nsw_dw_cpscontrol['ht'] = nsw_dw_cpscontrol['y1'] - nsw_dw_cpscontrol['y0']\n",
+    "\n",
+    "te_1 = nsw_dw_cpscontrol.ht.mean()\n",
+    "\n",
+    "print(\"Treatment Effect (non-normalized, all data): {:.2f}\".format(te_1))\n",
+    "        \n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Treatment Effect (normalized, all data): -7044.80\n"
+     ]
+    }
+   ],
+   "source": [
+    "nsw_dw_cpscontrol['y1'] = nsw_dw_cpscontrol.treat * nsw_dw_cpscontrol.re78 / nsw_dw_cpscontrol.pscore\n",
+    "nsw_dw_cpscontrol['y1'] /= s1/N\n",
+    "nsw_dw_cpscontrol['y0'] = (1 - nsw_dw_cpscontrol.treat) * nsw_dw_cpscontrol.re78 / (1 - nsw_dw_cpscontrol.pscore)\n",
+    "nsw_dw_cpscontrol['y0'] /= s0/N\n",
+    "nsw_dw_cpscontrol['ht'] = nsw_dw_cpscontrol['y1'] - nsw_dw_cpscontrol['y0']\n",
+    "\n",
+    "te_2 = nsw_dw_cpscontrol.ht.mean()\n",
+    "\n",
+    "print(\"Treatment Effect (normalized, all data): {:.2f}\".format(te_2))\n",
+    "        "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Treatment Effect (non-normalized, trimmed data): 2486.22\n"
+     ]
+    }
+   ],
+   "source": [
+    "nsw_dw_trimmed = nsw_dw_cpscontrol.drop(['d1', 'd0', 'y1', 'y0'], axis=1)\n",
+    "nsw_dw_trimmed = nsw_dw_trimmed[nsw_dw_trimmed.pscore.between(.1, .9)]\n",
+    "N = nsw_dw_trimmed.shape[0]\n",
+    "\n",
+    "nsw_dw_trimmed['y1'] = nsw_dw_trimmed.treat * nsw_dw_trimmed.re78 / nsw_dw_trimmed.pscore\n",
+    "nsw_dw_trimmed['y0'] = (1 - nsw_dw_trimmed.treat) * nsw_dw_trimmed.re78 / (1 - nsw_dw_trimmed.pscore)\n",
+    "nsw_dw_trimmed['ht'] = nsw_dw_trimmed['y1'] - nsw_dw_trimmed['y0']\n",
+    "\n",
+    "te_3 = nsw_dw_trimmed.ht.mean()\n",
+    "\n",
+    "print(\"Treatment Effect (non-normalized, trimmed data): {:.2f}\".format(te_3))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Treatment Effect (normalized, trimmed data): 423.62\n"
+     ]
+    }
+   ],
+   "source": [
+    "nsw_dw_trimmed['y1'] = nsw_dw_trimmed.treat * nsw_dw_trimmed.re78 / nsw_dw_trimmed.pscore\n",
+    "nsw_dw_trimmed['y1'] /= s1/N\n",
+    "nsw_dw_trimmed['y0'] = (1 - nsw_dw_trimmed.treat) * nsw_dw_trimmed.re78 / (1 - nsw_dw_trimmed.pscore)\n",
+    "nsw_dw_trimmed['y0'] /= s0/N\n",
+    "nsw_dw_trimmed['ht'] = nsw_dw_trimmed['y1'] - nsw_dw_trimmed['y0']\n",
+    "\n",
+    "te_4 = nsw_dw_trimmed.ht.mean()\n",
+    "\n",
+    "print(\"Treatment Effect (normalized, trimmed data): {:.2f}\".format(te_4))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "- Explain the overlap condition in the context of these data.  How did we ensure that overlap held in the data? \n",
+    "- When we are using non-trimmed data, why is the treatment effect negative? (*hint:* it has to do with extreme probability scores)\n",
+    "- What does this imply about the challenges of using non-experimental data when estimating causal effects, and why is conditioning on a trimmed propensity score important?\n",
+    "\n",
+    "\n",
+    "## Nearest-Neighbor Matching [not available in python]\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Compare our results from nearest-neighbor matching to what we found using the experimental data, the simple difference in outcomes using non-experimental controls, and propensity score weighting using non-experimental controls.\n",
+    "- DIFFICULT: Write a program that performs bootstrapping to get an estimate of the variance of the estimator. (HINT: Write a loop)\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  },
+  "toc": {
+   "base_numbering": 1,
+   "nav_menu": {},
+   "number_sections": true,
+   "sideBar": true,
+   "skip_h1_title": false,
+   "title_cell": "Table of Contents",
+   "title_sidebar": "Contents",
+   "toc_cell": false,
+   "toc_position": {},
+   "toc_section_display": true,
+   "toc_window_display": false
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/Python/Panel_Data.ipynb b/Python/Panel_Data.ipynb
new file mode 100644
index 0000000..00ee626
--- /dev/null
+++ b/Python/Panel_Data.ipynb
@@ -0,0 +1,2603 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Panel Data** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "from linearmodels import PanelOLS\n",
+    "import plotnine as p"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_stata(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sasp = read_data(\"sasp_panel.dta\")\n",
+    "#-- Delete all NA\n",
+    "sasp = sasp.dropna()\n",
+    "\n",
+    "#-- order by id and session \n",
+    "sasp.sort_values('id', inplace=True)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(1028, 32)"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#Balance Data\n",
+    "times = len(sasp.session.unique())\n",
+    "in_all_times = sasp.groupby('id')['session'].apply(lambda x : len(x)==times).reset_index()\n",
+    "in_all_times.rename(columns={'session':'in_all_times'}, inplace=True)\n",
+    "balanced_sasp = pd.merge(in_all_times, sasp, how='left', on='id')\n",
+    "balanced_sasp = balanced_sasp[balanced_sasp.in_all_times]\n",
+    "balanced_sasp.shape"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "provider_second = np.zeros(balanced_sasp.shape[0])\n",
+    "provider_second[balanced_sasp.provider_second == \"2. Yes\"] = 1\n",
+    "balanced_sasp.provider_second = provider_second"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Demean Data\n",
+    "features = balanced_sasp.columns.to_list()\n",
+    "features = [x for x in features if x not in ['session', 'id', 'in_all_times']]\n",
+    "demean_features = [\"demean_{}\".format(x) for x in features]\n",
+    "balanced_sasp[demean_features] = balanced_sasp.groupby('id')[features].apply(lambda x : x - np.mean(x))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>id</th>\n",
+       "      <th>in_all_times</th>\n",
+       "      <th>session</th>\n",
+       "      <th>age</th>\n",
+       "      <th>age_cl</th>\n",
+       "      <th>appearance_cl</th>\n",
+       "      <th>bmi</th>\n",
+       "      <th>schooling</th>\n",
+       "      <th>asq_cl</th>\n",
+       "      <th>provider_second</th>\n",
+       "      <th>...</th>\n",
+       "      <th>demean_hispanic</th>\n",
+       "      <th>demean_other</th>\n",
+       "      <th>demean_white</th>\n",
+       "      <th>demean_asq</th>\n",
+       "      <th>demean_cohab</th>\n",
+       "      <th>demean_married</th>\n",
+       "      <th>demean_divorced</th>\n",
+       "      <th>demean_separated</th>\n",
+       "      <th>demean_nevermarried</th>\n",
+       "      <th>demean_widowed</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>2025.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>32.5</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>1056.25</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>21.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>441.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>8.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>25.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>22.886999</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1369.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1485</th>\n",
+       "      <td>684.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>28.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>27.435720</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1495</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>35.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1225.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1496</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1497</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>2025.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1498</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>35.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1225.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>1028 rows × 61 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         id  in_all_times  session   age  age_cl  appearance_cl        bmi  \\\n",
+       "3       6.0          True      3.0  29.0    45.0            4.0  30.893555   \n",
+       "4       6.0          True      1.0  29.0    32.5            6.0  30.893555   \n",
+       "5       6.0          True      2.0  29.0    30.0            8.0  30.893555   \n",
+       "6       6.0          True      4.0  29.0    21.0            6.0  30.893555   \n",
+       "9       8.0          True      3.0  25.0    37.0            5.0  22.886999   \n",
+       "...     ...           ...      ...   ...     ...            ...        ...   \n",
+       "1485  684.0          True      2.0  28.0    30.0            6.0  27.435720   \n",
+       "1495  690.0          True      3.0  37.0    35.0            5.0  19.366392   \n",
+       "1496  690.0          True      2.0  37.0    30.0            6.0  19.366392   \n",
+       "1497  690.0          True      4.0  37.0    45.0            8.0  19.366392   \n",
+       "1498  690.0          True      1.0  37.0    35.0            8.0  19.366392   \n",
+       "\n",
+       "      schooling   asq_cl  provider_second  ...  demean_hispanic  demean_other  \\\n",
+       "3          16.0  2025.00              0.0  ...              0.0           0.0   \n",
+       "4          16.0  1056.25              0.0  ...              0.0           0.0   \n",
+       "5          16.0   900.00              0.0  ...              0.0           0.0   \n",
+       "6          16.0   441.00              0.0  ...              0.0           0.0   \n",
+       "9          14.0  1369.00              0.0  ...              0.0           0.0   \n",
+       "...         ...      ...              ...  ...              ...           ...   \n",
+       "1485       12.0   900.00              0.0  ...              0.0           0.0   \n",
+       "1495       14.0  1225.00              0.0  ...              0.0           0.0   \n",
+       "1496       14.0   900.00              0.0  ...              0.0           0.0   \n",
+       "1497       14.0  2025.00              0.0  ...              0.0           0.0   \n",
+       "1498       14.0  1225.00              0.0  ...              0.0           0.0   \n",
+       "\n",
+       "      demean_white  demean_asq  demean_cohab  demean_married  demean_divorced  \\\n",
+       "3              0.0         0.0           0.0             0.0              0.0   \n",
+       "4              0.0         0.0           0.0             0.0              0.0   \n",
+       "5              0.0         0.0           0.0             0.0              0.0   \n",
+       "6              0.0         0.0           0.0             0.0              0.0   \n",
+       "9              0.0         0.0           0.0             0.0              0.0   \n",
+       "...            ...         ...           ...             ...              ...   \n",
+       "1485           0.0         0.0           0.0             0.0              0.0   \n",
+       "1495           0.0         0.0           0.0             0.0              0.0   \n",
+       "1496           0.0         0.0           0.0             0.0              0.0   \n",
+       "1497           0.0         0.0           0.0             0.0              0.0   \n",
+       "1498           0.0         0.0           0.0             0.0              0.0   \n",
+       "\n",
+       "      demean_separated  demean_nevermarried  demean_widowed  \n",
+       "3                  0.0                  0.0             0.0  \n",
+       "4                  0.0                  0.0             0.0  \n",
+       "5                  0.0                  0.0             0.0  \n",
+       "6                  0.0                  0.0             0.0  \n",
+       "9                  0.0                  0.0             0.0  \n",
+       "...                ...                  ...             ...  \n",
+       "1485               0.0                  0.0             0.0  \n",
+       "1495               0.0                  0.0             0.0  \n",
+       "1496               0.0                  0.0             0.0  \n",
+       "1497               0.0                  0.0             0.0  \n",
+       "1498               0.0                  0.0             0.0  \n",
+       "\n",
+       "[1028 rows x 61 columns]"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "balanced_sasp"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Pooled OLS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>lnw</td>       <th>  R-squared:         </th> <td>   0.303</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.285</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   17.39</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>3.97e-62</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:32:50</td>     <th>  Log-Likelihood:    </th> <td> -570.00</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  1028</td>      <th>  AIC:               </th> <td>   1192.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>  1002</td>      <th>  BIC:               </th> <td>   1320.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    25</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "         <td></td>            <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>       <td>    7.0627</td> <td>    0.316</td> <td>   22.385</td> <td> 0.000</td> <td>    6.444</td> <td>    7.682</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>age</th>             <td>    0.0028</td> <td>    0.012</td> <td>    0.235</td> <td> 0.814</td> <td>   -0.020</td> <td>    0.026</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asq</th>             <td>   -0.0001</td> <td>    0.000</td> <td>   -0.828</td> <td> 0.408</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>bmi</th>             <td>   -0.0217</td> <td>    0.002</td> <td>   -9.296</td> <td> 0.000</td> <td>   -0.026</td> <td>   -0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hispanic</th>        <td>   -0.2259</td> <td>    0.091</td> <td>   -2.472</td> <td> 0.014</td> <td>   -0.405</td> <td>   -0.047</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black</th>           <td>    0.0284</td> <td>    0.075</td> <td>    0.379</td> <td> 0.705</td> <td>   -0.119</td> <td>    0.175</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>other</th>           <td>   -0.1116</td> <td>    0.061</td> <td>   -1.838</td> <td> 0.066</td> <td>   -0.231</td> <td>    0.008</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asian</th>           <td>    0.0862</td> <td>    0.154</td> <td>    0.559</td> <td> 0.576</td> <td>   -0.216</td> <td>    0.389</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>schooling</th>       <td>    0.0198</td> <td>    0.010</td> <td>    1.997</td> <td> 0.046</td> <td>    0.000</td> <td>    0.039</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>cohab</th>           <td>   -0.0540</td> <td>    0.040</td> <td>   -1.347</td> <td> 0.178</td> <td>   -0.133</td> <td>    0.025</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>married</th>         <td>    0.0046</td> <td>    0.042</td> <td>    0.110</td> <td> 0.912</td> <td>   -0.078</td> <td>    0.087</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>divorced</th>        <td>   -0.0209</td> <td>    0.039</td> <td>   -0.539</td> <td> 0.590</td> <td>   -0.097</td> <td>    0.055</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>separated</th>       <td>   -0.0557</td> <td>    0.063</td> <td>   -0.881</td> <td> 0.378</td> <td>   -0.180</td> <td>    0.068</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>age_cl</th>          <td>   -0.0013</td> <td>    0.009</td> <td>   -0.154</td> <td> 0.877</td> <td>   -0.018</td> <td>    0.016</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>unsafe</th>          <td>    0.0134</td> <td>    0.028</td> <td>    0.473</td> <td> 0.636</td> <td>   -0.042</td> <td>    0.069</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>llength</th>         <td>   -0.3083</td> <td>    0.020</td> <td>  -15.517</td> <td> 0.000</td> <td>   -0.347</td> <td>   -0.269</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>reg</th>             <td>   -0.0470</td> <td>    0.028</td> <td>   -1.657</td> <td> 0.098</td> <td>   -0.103</td> <td>    0.009</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asq_cl</th>          <td> 4.401e-05</td> <td> 9.12e-05</td> <td>    0.482</td> <td> 0.630</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>appearance_cl</th>   <td>    0.0200</td> <td>    0.007</td> <td>    2.986</td> <td> 0.003</td> <td>    0.007</td> <td>    0.033</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>provider_second</th> <td>    0.0554</td> <td>    0.072</td> <td>    0.765</td> <td> 0.445</td> <td>   -0.087</td> <td>    0.197</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asian_cl</th>        <td>   -0.0135</td> <td>    0.059</td> <td>   -0.231</td> <td> 0.818</td> <td>   -0.129</td> <td>    0.102</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black_cl</th>        <td>    0.0919</td> <td>    0.063</td> <td>    1.465</td> <td> 0.143</td> <td>   -0.031</td> <td>    0.215</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hispanic_cl</th>     <td>    0.0517</td> <td>    0.076</td> <td>    0.677</td> <td> 0.499</td> <td>   -0.098</td> <td>    0.201</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>othrace_cl</th>      <td>    0.1558</td> <td>    0.081</td> <td>    1.929</td> <td> 0.054</td> <td>   -0.003</td> <td>    0.314</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hot</th>             <td>    0.1332</td> <td>    0.028</td> <td>    4.722</td> <td> 0.000</td> <td>    0.078</td> <td>    0.188</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>massage_cl</th>      <td>   -0.1338</td> <td>    0.029</td> <td>   -4.585</td> <td> 0.000</td> <td>   -0.191</td> <td>   -0.077</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>62.662</td> <th>  Durbin-Watson:     </th> <td>   1.095</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 115.620</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.425</td> <th>  Prob(JB):          </th> <td>7.82e-26</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 4.405</td> <th>  Cond. No.          </th> <td>6.49e+04</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 6.49e+04. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                    lnw   R-squared:                       0.303\n",
+       "Model:                            OLS   Adj. R-squared:                  0.285\n",
+       "Method:                 Least Squares   F-statistic:                     17.39\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           3.97e-62\n",
+       "Time:                        13:32:50   Log-Likelihood:                -570.00\n",
+       "No. Observations:                1028   AIC:                             1192.\n",
+       "Df Residuals:                    1002   BIC:                             1320.\n",
+       "Df Model:                          25                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "===================================================================================\n",
+       "                      coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "-----------------------------------------------------------------------------------\n",
+       "Intercept           7.0627      0.316     22.385      0.000       6.444       7.682\n",
+       "age                 0.0028      0.012      0.235      0.814      -0.020       0.026\n",
+       "asq                -0.0001      0.000     -0.828      0.408      -0.000       0.000\n",
+       "bmi                -0.0217      0.002     -9.296      0.000      -0.026      -0.017\n",
+       "hispanic           -0.2259      0.091     -2.472      0.014      -0.405      -0.047\n",
+       "black               0.0284      0.075      0.379      0.705      -0.119       0.175\n",
+       "other              -0.1116      0.061     -1.838      0.066      -0.231       0.008\n",
+       "asian               0.0862      0.154      0.559      0.576      -0.216       0.389\n",
+       "schooling           0.0198      0.010      1.997      0.046       0.000       0.039\n",
+       "cohab              -0.0540      0.040     -1.347      0.178      -0.133       0.025\n",
+       "married             0.0046      0.042      0.110      0.912      -0.078       0.087\n",
+       "divorced           -0.0209      0.039     -0.539      0.590      -0.097       0.055\n",
+       "separated          -0.0557      0.063     -0.881      0.378      -0.180       0.068\n",
+       "age_cl             -0.0013      0.009     -0.154      0.877      -0.018       0.016\n",
+       "unsafe              0.0134      0.028      0.473      0.636      -0.042       0.069\n",
+       "llength            -0.3083      0.020    -15.517      0.000      -0.347      -0.269\n",
+       "reg                -0.0470      0.028     -1.657      0.098      -0.103       0.009\n",
+       "asq_cl           4.401e-05   9.12e-05      0.482      0.630      -0.000       0.000\n",
+       "appearance_cl       0.0200      0.007      2.986      0.003       0.007       0.033\n",
+       "provider_second     0.0554      0.072      0.765      0.445      -0.087       0.197\n",
+       "asian_cl           -0.0135      0.059     -0.231      0.818      -0.129       0.102\n",
+       "black_cl            0.0919      0.063      1.465      0.143      -0.031       0.215\n",
+       "hispanic_cl         0.0517      0.076      0.677      0.499      -0.098       0.201\n",
+       "othrace_cl          0.1558      0.081      1.929      0.054      -0.003       0.314\n",
+       "hot                 0.1332      0.028      4.722      0.000       0.078       0.188\n",
+       "massage_cl         -0.1338      0.029     -4.585      0.000      -0.191      -0.077\n",
+       "==============================================================================\n",
+       "Omnibus:                       62.662   Durbin-Watson:                   1.095\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              115.620\n",
+       "Skew:                           0.425   Prob(JB):                     7.82e-26\n",
+       "Kurtosis:                       4.405   Cond. No.                     6.49e+04\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "[2] The condition number is large, 6.49e+04. This might indicate that there are\n",
+       "strong multicollinearity or other numerical problems.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "dep_var = \"+\".join(features)\n",
+    "formula = \"\"\"lnw ~ age + asq + bmi + hispanic + black + other + asian + schooling + cohab + \n",
+    "            married + divorced + separated + age_cl + unsafe + llength + reg + asq_cl + \n",
+    "            appearance_cl + provider_second + asian_cl + black_cl + hispanic_cl + \n",
+    "           othrace_cl + hot + massage_cl\"\"\"\n",
+    "ols = sm.OLS.from_formula(formula, data=balanced_sasp).fit()\n",
+    "ols.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Fixed Effects"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "balanced_sasp['y'] = balanced_sasp.lnw"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>id</th>\n",
+       "      <th>in_all_times</th>\n",
+       "      <th>session</th>\n",
+       "      <th>age</th>\n",
+       "      <th>age_cl</th>\n",
+       "      <th>appearance_cl</th>\n",
+       "      <th>bmi</th>\n",
+       "      <th>schooling</th>\n",
+       "      <th>asq_cl</th>\n",
+       "      <th>provider_second</th>\n",
+       "      <th>...</th>\n",
+       "      <th>demean_other</th>\n",
+       "      <th>demean_white</th>\n",
+       "      <th>demean_asq</th>\n",
+       "      <th>demean_cohab</th>\n",
+       "      <th>demean_married</th>\n",
+       "      <th>demean_divorced</th>\n",
+       "      <th>demean_separated</th>\n",
+       "      <th>demean_nevermarried</th>\n",
+       "      <th>demean_widowed</th>\n",
+       "      <th>y</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>2025.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.192957</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>32.5</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>1056.25</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.585999</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>5</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.585999</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>6</th>\n",
+       "      <td>6.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>29.0</td>\n",
+       "      <td>21.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>30.893555</td>\n",
+       "      <td>16.0</td>\n",
+       "      <td>441.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.010635</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>9</th>\n",
+       "      <td>8.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>25.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>22.886999</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1369.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>4.605170</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1485</th>\n",
+       "      <td>684.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>28.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>27.435720</td>\n",
+       "      <td>12.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>6.396930</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1495</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>3.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>35.0</td>\n",
+       "      <td>5.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1225.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.298317</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1496</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>2.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>30.0</td>\n",
+       "      <td>6.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>900.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.298317</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1497</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>4.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>45.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>2025.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.991465</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1498</th>\n",
+       "      <td>690.0</td>\n",
+       "      <td>True</td>\n",
+       "      <td>1.0</td>\n",
+       "      <td>37.0</td>\n",
+       "      <td>35.0</td>\n",
+       "      <td>8.0</td>\n",
+       "      <td>19.366392</td>\n",
+       "      <td>14.0</td>\n",
+       "      <td>1225.00</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>...</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>0.0</td>\n",
+       "      <td>5.298317</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>1028 rows × 62 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "         id  in_all_times  session   age  age_cl  appearance_cl        bmi  \\\n",
+       "3       6.0          True      3.0  29.0    45.0            4.0  30.893555   \n",
+       "4       6.0          True      1.0  29.0    32.5            6.0  30.893555   \n",
+       "5       6.0          True      2.0  29.0    30.0            8.0  30.893555   \n",
+       "6       6.0          True      4.0  29.0    21.0            6.0  30.893555   \n",
+       "9       8.0          True      3.0  25.0    37.0            5.0  22.886999   \n",
+       "...     ...           ...      ...   ...     ...            ...        ...   \n",
+       "1485  684.0          True      2.0  28.0    30.0            6.0  27.435720   \n",
+       "1495  690.0          True      3.0  37.0    35.0            5.0  19.366392   \n",
+       "1496  690.0          True      2.0  37.0    30.0            6.0  19.366392   \n",
+       "1497  690.0          True      4.0  37.0    45.0            8.0  19.366392   \n",
+       "1498  690.0          True      1.0  37.0    35.0            8.0  19.366392   \n",
+       "\n",
+       "      schooling   asq_cl  provider_second  ...  demean_other  demean_white  \\\n",
+       "3          16.0  2025.00              0.0  ...           0.0           0.0   \n",
+       "4          16.0  1056.25              0.0  ...           0.0           0.0   \n",
+       "5          16.0   900.00              0.0  ...           0.0           0.0   \n",
+       "6          16.0   441.00              0.0  ...           0.0           0.0   \n",
+       "9          14.0  1369.00              0.0  ...           0.0           0.0   \n",
+       "...         ...      ...              ...  ...           ...           ...   \n",
+       "1485       12.0   900.00              0.0  ...           0.0           0.0   \n",
+       "1495       14.0  1225.00              0.0  ...           0.0           0.0   \n",
+       "1496       14.0   900.00              0.0  ...           0.0           0.0   \n",
+       "1497       14.0  2025.00              0.0  ...           0.0           0.0   \n",
+       "1498       14.0  1225.00              0.0  ...           0.0           0.0   \n",
+       "\n",
+       "      demean_asq  demean_cohab  demean_married  demean_divorced  \\\n",
+       "3            0.0           0.0             0.0              0.0   \n",
+       "4            0.0           0.0             0.0              0.0   \n",
+       "5            0.0           0.0             0.0              0.0   \n",
+       "6            0.0           0.0             0.0              0.0   \n",
+       "9            0.0           0.0             0.0              0.0   \n",
+       "...          ...           ...             ...              ...   \n",
+       "1485         0.0           0.0             0.0              0.0   \n",
+       "1495         0.0           0.0             0.0              0.0   \n",
+       "1496         0.0           0.0             0.0              0.0   \n",
+       "1497         0.0           0.0             0.0              0.0   \n",
+       "1498         0.0           0.0             0.0              0.0   \n",
+       "\n",
+       "      demean_separated  demean_nevermarried  demean_widowed         y  \n",
+       "3                  0.0                  0.0             0.0  5.192957  \n",
+       "4                  0.0                  0.0             0.0  5.585999  \n",
+       "5                  0.0                  0.0             0.0  5.585999  \n",
+       "6                  0.0                  0.0             0.0  5.010635  \n",
+       "9                  0.0                  0.0             0.0  4.605170  \n",
+       "...                ...                  ...             ...       ...  \n",
+       "1485               0.0                  0.0             0.0  6.396930  \n",
+       "1495               0.0                  0.0             0.0  5.298317  \n",
+       "1496               0.0                  0.0             0.0  5.298317  \n",
+       "1497               0.0                  0.0             0.0  5.991465  \n",
+       "1498               0.0                  0.0             0.0  5.298317  \n",
+       "\n",
+       "[1028 rows x 62 columns]"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "balanced_sasp"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>lnw</td>       <th>  R-squared:         </th> <td>   0.832</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.773</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>     nan</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th>  <td>   nan</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:32:50</td>     <th>  Log-Likelihood:    </th> <td>  162.25</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  1028</td>      <th>  AIC:               </th> <td>   215.5</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   758</td>      <th>  BIC:               </th> <td>   1548.</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>   269</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "         <td></td>            <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[6.0]</th>      <td>   -0.4465</td> <td>    0.035</td> <td>  -12.683</td> <td> 0.000</td> <td>   -0.515</td> <td>   -0.377</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[8.0]</th>      <td>    0.3310</td> <td>    0.027</td> <td>   12.240</td> <td> 0.000</td> <td>    0.278</td> <td>    0.384</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[10.0]</th>     <td>    1.0513</td> <td>    0.040</td> <td>   26.539</td> <td> 0.000</td> <td>    0.974</td> <td>    1.129</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[11.0]</th>     <td>   -0.5627</td> <td>    0.026</td> <td>  -21.608</td> <td> 0.000</td> <td>   -0.614</td> <td>   -0.512</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[18.0]</th>     <td>    0.5518</td> <td>    0.034</td> <td>   16.025</td> <td> 0.000</td> <td>    0.484</td> <td>    0.619</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[23.0]</th>     <td>   -0.0312</td> <td>    0.038</td> <td>   -0.827</td> <td> 0.408</td> <td>   -0.105</td> <td>    0.043</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[25.0]</th>     <td>   -0.1525</td> <td>    0.035</td> <td>   -4.403</td> <td> 0.000</td> <td>   -0.220</td> <td>   -0.085</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[29.0]</th>     <td>    1.6517</td> <td>    0.068</td> <td>   24.285</td> <td> 0.000</td> <td>    1.518</td> <td>    1.785</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[31.0]</th>     <td>   -0.0586</td> <td>    0.020</td> <td>   -2.953</td> <td> 0.003</td> <td>   -0.098</td> <td>   -0.020</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[33.0]</th>     <td>    0.3145</td> <td>    0.021</td> <td>   14.914</td> <td> 0.000</td> <td>    0.273</td> <td>    0.356</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[34.0]</th>     <td>   -0.4864</td> <td>    0.030</td> <td>  -16.439</td> <td> 0.000</td> <td>   -0.544</td> <td>   -0.428</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[35.0]</th>     <td>    0.8514</td> <td>    0.040</td> <td>   21.275</td> <td> 0.000</td> <td>    0.773</td> <td>    0.930</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[42.0]</th>     <td>    0.8090</td> <td>    0.040</td> <td>   20.387</td> <td> 0.000</td> <td>    0.731</td> <td>    0.887</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[44.0]</th>     <td>   -0.3047</td> <td>    0.035</td> <td>   -8.649</td> <td> 0.000</td> <td>   -0.374</td> <td>   -0.236</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[48.0]</th>     <td>   -0.3631</td> <td>    0.030</td> <td>  -12.024</td> <td> 0.000</td> <td>   -0.422</td> <td>   -0.304</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[53.0]</th>     <td>    0.3511</td> <td>    0.031</td> <td>   11.184</td> <td> 0.000</td> <td>    0.290</td> <td>    0.413</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[54.0]</th>     <td>    0.6716</td> <td>    0.036</td> <td>   18.840</td> <td> 0.000</td> <td>    0.602</td> <td>    0.741</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[56.0]</th>     <td>   -0.4843</td> <td>    0.024</td> <td>  -20.443</td> <td> 0.000</td> <td>   -0.531</td> <td>   -0.438</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[59.0]</th>     <td>   -0.3581</td> <td>    0.025</td> <td>  -14.086</td> <td> 0.000</td> <td>   -0.408</td> <td>   -0.308</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[61.0]</th>     <td>   -0.4522</td> <td>    0.057</td> <td>   -7.929</td> <td> 0.000</td> <td>   -0.564</td> <td>   -0.340</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[62.0]</th>     <td>   -0.2078</td> <td>    0.033</td> <td>   -6.232</td> <td> 0.000</td> <td>   -0.273</td> <td>   -0.142</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[64.0]</th>     <td>   -0.3638</td> <td>    0.038</td> <td>   -9.589</td> <td> 0.000</td> <td>   -0.438</td> <td>   -0.289</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[68.0]</th>     <td>    0.2320</td> <td>    0.028</td> <td>    8.163</td> <td> 0.000</td> <td>    0.176</td> <td>    0.288</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[72.0]</th>     <td>    0.0063</td> <td>    0.033</td> <td>    0.191</td> <td> 0.849</td> <td>   -0.058</td> <td>    0.070</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[73.0]</th>     <td>   -0.1538</td> <td>    0.030</td> <td>   -5.065</td> <td> 0.000</td> <td>   -0.213</td> <td>   -0.094</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[77.0]</th>     <td>   -0.4928</td> <td>    0.037</td> <td>  -13.168</td> <td> 0.000</td> <td>   -0.566</td> <td>   -0.419</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[78.0]</th>     <td>   -0.5062</td> <td>    0.034</td> <td>  -14.938</td> <td> 0.000</td> <td>   -0.573</td> <td>   -0.440</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[79.0]</th>     <td>    0.9076</td> <td>    0.044</td> <td>   20.816</td> <td> 0.000</td> <td>    0.822</td> <td>    0.993</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[83.0]</th>     <td>   -0.4709</td> <td>    0.015</td> <td>  -30.601</td> <td> 0.000</td> <td>   -0.501</td> <td>   -0.441</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[84.0]</th>     <td>   -0.8525</td> <td>    0.037</td> <td>  -22.855</td> <td> 0.000</td> <td>   -0.926</td> <td>   -0.779</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[86.0]</th>     <td>    0.7815</td> <td>    0.036</td> <td>   21.836</td> <td> 0.000</td> <td>    0.711</td> <td>    0.852</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[88.0]</th>     <td>    0.5577</td> <td>    0.031</td> <td>   17.889</td> <td> 0.000</td> <td>    0.497</td> <td>    0.619</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[91.0]</th>     <td>   -0.7363</td> <td>    0.043</td> <td>  -16.985</td> <td> 0.000</td> <td>   -0.821</td> <td>   -0.651</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[93.0]</th>     <td>    0.2723</td> <td>    0.029</td> <td>    9.354</td> <td> 0.000</td> <td>    0.215</td> <td>    0.329</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[94.0]</th>     <td>    0.5675</td> <td>    0.050</td> <td>   11.372</td> <td> 0.000</td> <td>    0.470</td> <td>    0.665</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[96.0]</th>     <td>    0.5108</td> <td>    0.026</td> <td>   19.505</td> <td> 0.000</td> <td>    0.459</td> <td>    0.562</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[100.0]</th>    <td>   -0.2375</td> <td>    0.041</td> <td>   -5.815</td> <td> 0.000</td> <td>   -0.318</td> <td>   -0.157</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[105.0]</th>    <td>    0.6224</td> <td>    0.031</td> <td>   19.909</td> <td> 0.000</td> <td>    0.561</td> <td>    0.684</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[107.0]</th>    <td>   -0.8367</td> <td>    0.041</td> <td>  -20.566</td> <td> 0.000</td> <td>   -0.916</td> <td>   -0.757</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[108.0]</th>    <td>   -0.4168</td> <td>    0.050</td> <td>   -8.403</td> <td> 0.000</td> <td>   -0.514</td> <td>   -0.320</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[113.0]</th>    <td>    0.0119</td> <td>    0.020</td> <td>    0.582</td> <td> 0.561</td> <td>   -0.028</td> <td>    0.052</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[114.0]</th>    <td>   -0.1436</td> <td>    0.028</td> <td>   -5.102</td> <td> 0.000</td> <td>   -0.199</td> <td>   -0.088</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[115.0]</th>    <td>    0.2745</td> <td>    0.042</td> <td>    6.474</td> <td> 0.000</td> <td>    0.191</td> <td>    0.358</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[120.0]</th>    <td>   -0.0830</td> <td>    0.025</td> <td>   -3.334</td> <td> 0.001</td> <td>   -0.132</td> <td>   -0.034</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[121.0]</th>    <td>    0.0400</td> <td>    0.042</td> <td>    0.940</td> <td> 0.347</td> <td>   -0.043</td> <td>    0.123</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[122.0]</th>    <td>    0.2025</td> <td>    0.030</td> <td>    6.683</td> <td> 0.000</td> <td>    0.143</td> <td>    0.262</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[123.0]</th>    <td>    0.1032</td> <td>    0.012</td> <td>    8.494</td> <td> 0.000</td> <td>    0.079</td> <td>    0.127</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[128.0]</th>    <td>    0.1970</td> <td>    0.027</td> <td>    7.224</td> <td> 0.000</td> <td>    0.144</td> <td>    0.250</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[130.0]</th>    <td>   -0.8677</td> <td>    0.034</td> <td>  -25.749</td> <td> 0.000</td> <td>   -0.934</td> <td>   -0.802</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[131.0]</th>    <td>   -1.0605</td> <td>    0.040</td> <td>  -26.442</td> <td> 0.000</td> <td>   -1.139</td> <td>   -0.982</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[134.0]</th>    <td>   -0.3681</td> <td>    0.031</td> <td>  -11.754</td> <td> 0.000</td> <td>   -0.429</td> <td>   -0.307</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[137.0]</th>    <td>   -1.1538</td> <td>    0.043</td> <td>  -26.747</td> <td> 0.000</td> <td>   -1.238</td> <td>   -1.069</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[138.0]</th>    <td>    0.8507</td> <td>    0.027</td> <td>   31.608</td> <td> 0.000</td> <td>    0.798</td> <td>    0.903</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[140.0]</th>    <td>   -0.7737</td> <td>    0.040</td> <td>  -19.503</td> <td> 0.000</td> <td>   -0.851</td> <td>   -0.696</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[145.0]</th>    <td>   -0.5292</td> <td>    0.033</td> <td>  -16.207</td> <td> 0.000</td> <td>   -0.593</td> <td>   -0.465</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[148.0]</th>    <td>    0.7798</td> <td>    0.036</td> <td>   21.739</td> <td> 0.000</td> <td>    0.709</td> <td>    0.850</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[149.0]</th>    <td>    0.4930</td> <td>    0.033</td> <td>   15.156</td> <td> 0.000</td> <td>    0.429</td> <td>    0.557</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[150.0]</th>    <td>    1.3854</td> <td>    0.048</td> <td>   29.121</td> <td> 0.000</td> <td>    1.292</td> <td>    1.479</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[151.0]</th>    <td>   -0.1263</td> <td>    0.030</td> <td>   -4.144</td> <td> 0.000</td> <td>   -0.186</td> <td>   -0.067</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[153.0]</th>    <td>   -0.3541</td> <td>    0.034</td> <td>  -10.319</td> <td> 0.000</td> <td>   -0.421</td> <td>   -0.287</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[156.0]</th>    <td>    0.1443</td> <td>    0.035</td> <td>    4.113</td> <td> 0.000</td> <td>    0.076</td> <td>    0.213</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[158.0]</th>    <td>   -0.5936</td> <td>    0.022</td> <td>  -26.833</td> <td> 0.000</td> <td>   -0.637</td> <td>   -0.550</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[159.0]</th>    <td>   -0.1781</td> <td>    0.039</td> <td>   -4.525</td> <td> 0.000</td> <td>   -0.255</td> <td>   -0.101</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[164.0]</th>    <td>    0.1872</td> <td>    0.031</td> <td>    6.136</td> <td> 0.000</td> <td>    0.127</td> <td>    0.247</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[165.0]</th>    <td>    0.7737</td> <td>    0.029</td> <td>   26.643</td> <td> 0.000</td> <td>    0.717</td> <td>    0.831</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[166.0]</th>    <td>   -0.2982</td> <td>    0.025</td> <td>  -12.043</td> <td> 0.000</td> <td>   -0.347</td> <td>   -0.250</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[168.0]</th>    <td>   -0.9939</td> <td>    0.046</td> <td>  -21.529</td> <td> 0.000</td> <td>   -1.084</td> <td>   -0.903</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[170.0]</th>    <td>    0.4943</td> <td>    0.025</td> <td>   19.974</td> <td> 0.000</td> <td>    0.446</td> <td>    0.543</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[173.0]</th>    <td>    0.7285</td> <td>    0.027</td> <td>   27.433</td> <td> 0.000</td> <td>    0.676</td> <td>    0.781</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[176.0]</th>    <td>   -0.3965</td> <td>    0.027</td> <td>  -14.515</td> <td> 0.000</td> <td>   -0.450</td> <td>   -0.343</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[177.0]</th>    <td>    0.0037</td> <td>    0.027</td> <td>    0.136</td> <td> 0.892</td> <td>   -0.049</td> <td>    0.057</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[179.0]</th>    <td>   -0.7709</td> <td>    0.035</td> <td>  -22.034</td> <td> 0.000</td> <td>   -0.840</td> <td>   -0.702</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[180.0]</th>    <td>   -0.2945</td> <td>    0.033</td> <td>   -8.922</td> <td> 0.000</td> <td>   -0.359</td> <td>   -0.230</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[181.0]</th>    <td>    0.2046</td> <td>    0.024</td> <td>    8.695</td> <td> 0.000</td> <td>    0.159</td> <td>    0.251</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[183.0]</th>    <td>   -0.0737</td> <td>    0.026</td> <td>   -2.785</td> <td> 0.005</td> <td>   -0.126</td> <td>   -0.022</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[185.0]</th>    <td>    0.7862</td> <td>    0.041</td> <td>   19.218</td> <td> 0.000</td> <td>    0.706</td> <td>    0.866</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[186.0]</th>    <td>   -0.3200</td> <td>    0.033</td> <td>   -9.666</td> <td> 0.000</td> <td>   -0.385</td> <td>   -0.255</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[188.0]</th>    <td>    0.4196</td> <td>    0.041</td> <td>   10.346</td> <td> 0.000</td> <td>    0.340</td> <td>    0.499</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[189.0]</th>    <td>   -0.3912</td> <td>    0.055</td> <td>   -7.150</td> <td> 0.000</td> <td>   -0.498</td> <td>   -0.284</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[191.0]</th>    <td>    0.0024</td> <td>    0.036</td> <td>    0.067</td> <td> 0.947</td> <td>   -0.069</td> <td>    0.074</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[193.0]</th>    <td>   -0.7278</td> <td>    0.042</td> <td>  -17.355</td> <td> 0.000</td> <td>   -0.810</td> <td>   -0.646</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[195.0]</th>    <td>    0.4724</td> <td>    0.031</td> <td>   15.059</td> <td> 0.000</td> <td>    0.411</td> <td>    0.534</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[199.0]</th>    <td>   -0.8182</td> <td>    0.022</td> <td>  -37.782</td> <td> 0.000</td> <td>   -0.861</td> <td>   -0.776</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[200.0]</th>    <td>    0.1035</td> <td>    0.028</td> <td>    3.701</td> <td> 0.000</td> <td>    0.049</td> <td>    0.158</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[202.0]</th>    <td>    0.3100</td> <td>    0.032</td> <td>    9.700</td> <td> 0.000</td> <td>    0.247</td> <td>    0.373</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[206.0]</th>    <td>    0.2463</td> <td>    0.028</td> <td>    8.652</td> <td> 0.000</td> <td>    0.190</td> <td>    0.302</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[215.0]</th>    <td>   -0.3703</td> <td>    0.026</td> <td>  -14.404</td> <td> 0.000</td> <td>   -0.421</td> <td>   -0.320</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[220.0]</th>    <td>   -0.0165</td> <td>    0.028</td> <td>   -0.581</td> <td> 0.561</td> <td>   -0.072</td> <td>    0.039</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[223.0]</th>    <td>    0.3535</td> <td>    0.041</td> <td>    8.533</td> <td> 0.000</td> <td>    0.272</td> <td>    0.435</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[225.0]</th>    <td>    0.1590</td> <td>    0.019</td> <td>    8.565</td> <td> 0.000</td> <td>    0.123</td> <td>    0.195</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[227.0]</th>    <td>    0.1842</td> <td>    0.034</td> <td>    5.361</td> <td> 0.000</td> <td>    0.117</td> <td>    0.252</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[229.0]</th>    <td>   -0.3853</td> <td>    0.023</td> <td>  -17.076</td> <td> 0.000</td> <td>   -0.429</td> <td>   -0.341</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[230.0]</th>    <td>    0.3953</td> <td>    0.032</td> <td>   12.419</td> <td> 0.000</td> <td>    0.333</td> <td>    0.458</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[231.0]</th>    <td>    0.0641</td> <td>    0.034</td> <td>    1.874</td> <td> 0.061</td> <td>   -0.003</td> <td>    0.131</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[238.0]</th>    <td>    0.1458</td> <td>    0.050</td> <td>    2.906</td> <td> 0.004</td> <td>    0.047</td> <td>    0.244</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[239.0]</th>    <td>   -0.4659</td> <td>    0.045</td> <td>  -10.334</td> <td> 0.000</td> <td>   -0.554</td> <td>   -0.378</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[246.0]</th>    <td>   -0.0273</td> <td>    0.023</td> <td>   -1.199</td> <td> 0.231</td> <td>   -0.072</td> <td>    0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[253.0]</th>    <td>   -0.7272</td> <td>    0.039</td> <td>  -18.829</td> <td> 0.000</td> <td>   -0.803</td> <td>   -0.651</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[257.0]</th>    <td>   -0.1557</td> <td>    0.027</td> <td>   -5.671</td> <td> 0.000</td> <td>   -0.210</td> <td>   -0.102</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[271.0]</th>    <td>   -0.2892</td> <td>    0.021</td> <td>  -14.042</td> <td> 0.000</td> <td>   -0.330</td> <td>   -0.249</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[274.0]</th>    <td>    0.2065</td> <td>    0.038</td> <td>    5.448</td> <td> 0.000</td> <td>    0.132</td> <td>    0.281</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[279.0]</th>    <td>   -0.5520</td> <td>    0.037</td> <td>  -14.896</td> <td> 0.000</td> <td>   -0.625</td> <td>   -0.479</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[280.0]</th>    <td>    0.4347</td> <td>    0.032</td> <td>   13.642</td> <td> 0.000</td> <td>    0.372</td> <td>    0.497</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[281.0]</th>    <td>    0.4606</td> <td>    0.040</td> <td>   11.514</td> <td> 0.000</td> <td>    0.382</td> <td>    0.539</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[282.0]</th>    <td>   -0.0460</td> <td>    0.023</td> <td>   -2.031</td> <td> 0.042</td> <td>   -0.090</td> <td>   -0.002</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[283.0]</th>    <td>   -0.1643</td> <td>    0.024</td> <td>   -6.935</td> <td> 0.000</td> <td>   -0.211</td> <td>   -0.118</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[292.0]</th>    <td>    0.3708</td> <td>    0.040</td> <td>    9.307</td> <td> 0.000</td> <td>    0.293</td> <td>    0.449</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[295.0]</th>    <td>    1.4466</td> <td>    0.028</td> <td>   51.492</td> <td> 0.000</td> <td>    1.391</td> <td>    1.502</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[298.0]</th>    <td>    1.3459</td> <td>    0.042</td> <td>   32.059</td> <td> 0.000</td> <td>    1.264</td> <td>    1.428</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[299.0]</th>    <td>    0.2884</td> <td>    0.033</td> <td>    8.680</td> <td> 0.000</td> <td>    0.223</td> <td>    0.354</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[303.0]</th>    <td>   -0.6055</td> <td>    0.029</td> <td>  -20.985</td> <td> 0.000</td> <td>   -0.662</td> <td>   -0.549</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[304.0]</th>    <td>    0.0778</td> <td>    0.029</td> <td>    2.642</td> <td> 0.008</td> <td>    0.020</td> <td>    0.136</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[306.0]</th>    <td>    0.5709</td> <td>    0.031</td> <td>   18.691</td> <td> 0.000</td> <td>    0.511</td> <td>    0.631</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[308.0]</th>    <td>    0.3815</td> <td>    0.028</td> <td>   13.443</td> <td> 0.000</td> <td>    0.326</td> <td>    0.437</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[309.0]</th>    <td>    0.0374</td> <td>    0.029</td> <td>    1.301</td> <td> 0.193</td> <td>   -0.019</td> <td>    0.094</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[310.0]</th>    <td>    1.3725</td> <td>    0.039</td> <td>   34.955</td> <td> 0.000</td> <td>    1.296</td> <td>    1.449</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[314.0]</th>    <td>   -0.3016</td> <td>    0.031</td> <td>   -9.730</td> <td> 0.000</td> <td>   -0.362</td> <td>   -0.241</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[315.0]</th>    <td>   -0.1683</td> <td>    0.033</td> <td>   -5.028</td> <td> 0.000</td> <td>   -0.234</td> <td>   -0.103</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[334.0]</th>    <td>   -0.1043</td> <td>    0.029</td> <td>   -3.542</td> <td> 0.000</td> <td>   -0.162</td> <td>   -0.047</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[335.0]</th>    <td>    0.3782</td> <td>    0.021</td> <td>   18.014</td> <td> 0.000</td> <td>    0.337</td> <td>    0.419</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[337.0]</th>    <td>   -0.9128</td> <td>    0.029</td> <td>  -31.659</td> <td> 0.000</td> <td>   -0.969</td> <td>   -0.856</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[339.0]</th>    <td>   -0.3775</td> <td>    0.037</td> <td>  -10.266</td> <td> 0.000</td> <td>   -0.450</td> <td>   -0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[340.0]</th>    <td>    0.6570</td> <td>    0.043</td> <td>   15.280</td> <td> 0.000</td> <td>    0.573</td> <td>    0.741</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[343.0]</th>    <td>   -0.1178</td> <td>    0.028</td> <td>   -4.168</td> <td> 0.000</td> <td>   -0.173</td> <td>   -0.062</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[345.0]</th>    <td>    1.4559</td> <td>    0.029</td> <td>   50.267</td> <td> 0.000</td> <td>    1.399</td> <td>    1.513</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[351.0]</th>    <td>   -0.3798</td> <td>    0.042</td> <td>   -8.998</td> <td> 0.000</td> <td>   -0.463</td> <td>   -0.297</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[357.0]</th>    <td>    0.1202</td> <td>    0.035</td> <td>    3.429</td> <td> 0.001</td> <td>    0.051</td> <td>    0.189</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[358.0]</th>    <td>    0.3197</td> <td>    0.032</td> <td>    9.853</td> <td> 0.000</td> <td>    0.256</td> <td>    0.383</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[361.0]</th>    <td>   -0.2092</td> <td>    0.042</td> <td>   -4.970</td> <td> 0.000</td> <td>   -0.292</td> <td>   -0.127</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[364.0]</th>    <td>   -0.2560</td> <td>    0.055</td> <td>   -4.632</td> <td> 0.000</td> <td>   -0.364</td> <td>   -0.148</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[365.0]</th>    <td>   -0.3782</td> <td>    0.026</td> <td>  -14.416</td> <td> 0.000</td> <td>   -0.430</td> <td>   -0.327</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[368.0]</th>    <td>   -0.3443</td> <td>    0.037</td> <td>   -9.335</td> <td> 0.000</td> <td>   -0.417</td> <td>   -0.272</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[370.0]</th>    <td>   -0.0859</td> <td>    0.022</td> <td>   -3.933</td> <td> 0.000</td> <td>   -0.129</td> <td>   -0.043</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[371.0]</th>    <td>    0.7179</td> <td>    0.041</td> <td>   17.335</td> <td> 0.000</td> <td>    0.637</td> <td>    0.799</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[373.0]</th>    <td>    0.2087</td> <td>    0.023</td> <td>    8.924</td> <td> 0.000</td> <td>    0.163</td> <td>    0.255</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[376.0]</th>    <td>   -0.5252</td> <td>    0.038</td> <td>  -13.673</td> <td> 0.000</td> <td>   -0.600</td> <td>   -0.450</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[378.0]</th>    <td>    0.2573</td> <td>    0.033</td> <td>    7.689</td> <td> 0.000</td> <td>    0.192</td> <td>    0.323</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[381.0]</th>    <td>    1.0624</td> <td>    0.026</td> <td>   40.503</td> <td> 0.000</td> <td>    1.011</td> <td>    1.114</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[382.0]</th>    <td>   -0.8283</td> <td>    0.042</td> <td>  -19.778</td> <td> 0.000</td> <td>   -0.910</td> <td>   -0.746</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[388.0]</th>    <td>   -0.7960</td> <td>    0.038</td> <td>  -21.018</td> <td> 0.000</td> <td>   -0.870</td> <td>   -0.722</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[393.0]</th>    <td>   -0.5925</td> <td>    0.035</td> <td>  -16.706</td> <td> 0.000</td> <td>   -0.662</td> <td>   -0.523</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[394.0]</th>    <td>   -0.5698</td> <td>    0.029</td> <td>  -19.550</td> <td> 0.000</td> <td>   -0.627</td> <td>   -0.513</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[395.0]</th>    <td>   -0.0093</td> <td>    0.038</td> <td>   -0.244</td> <td> 0.807</td> <td>   -0.084</td> <td>    0.065</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[396.0]</th>    <td>   -0.9476</td> <td>    0.042</td> <td>  -22.397</td> <td> 0.000</td> <td>   -1.031</td> <td>   -0.865</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[397.0]</th>    <td>    0.0325</td> <td>    0.027</td> <td>    1.227</td> <td> 0.220</td> <td>   -0.019</td> <td>    0.085</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[403.0]</th>    <td>   -0.1233</td> <td>    0.032</td> <td>   -3.848</td> <td> 0.000</td> <td>   -0.186</td> <td>   -0.060</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[404.0]</th>    <td>    0.1958</td> <td>    0.029</td> <td>    6.853</td> <td> 0.000</td> <td>    0.140</td> <td>    0.252</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[410.0]</th>    <td>   -0.3268</td> <td>    0.011</td> <td>  -29.689</td> <td> 0.000</td> <td>   -0.348</td> <td>   -0.305</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[416.0]</th>    <td>    0.0475</td> <td>    0.031</td> <td>    1.518</td> <td> 0.129</td> <td>   -0.014</td> <td>    0.109</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[421.0]</th>    <td>    0.6134</td> <td>    0.029</td> <td>   21.068</td> <td> 0.000</td> <td>    0.556</td> <td>    0.670</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[422.0]</th>    <td>    0.3976</td> <td>    0.043</td> <td>    9.272</td> <td> 0.000</td> <td>    0.314</td> <td>    0.482</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[424.0]</th>    <td>   -0.7169</td> <td>    0.028</td> <td>  -25.183</td> <td> 0.000</td> <td>   -0.773</td> <td>   -0.661</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[427.0]</th>    <td>   -0.7037</td> <td>    0.030</td> <td>  -23.804</td> <td> 0.000</td> <td>   -0.762</td> <td>   -0.646</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[432.0]</th>    <td>   -0.9292</td> <td>    0.033</td> <td>  -28.090</td> <td> 0.000</td> <td>   -0.994</td> <td>   -0.864</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[433.0]</th>    <td>    0.1764</td> <td>    0.042</td> <td>    4.169</td> <td> 0.000</td> <td>    0.093</td> <td>    0.259</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[440.0]</th>    <td>    0.2746</td> <td>    0.042</td> <td>    6.583</td> <td> 0.000</td> <td>    0.193</td> <td>    0.356</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[441.0]</th>    <td>    0.2780</td> <td>    0.028</td> <td>    9.806</td> <td> 0.000</td> <td>    0.222</td> <td>    0.334</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[443.0]</th>    <td>   -0.3816</td> <td>    0.017</td> <td>  -22.708</td> <td> 0.000</td> <td>   -0.415</td> <td>   -0.349</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[445.0]</th>    <td>   -0.5643</td> <td>    0.037</td> <td>  -15.090</td> <td> 0.000</td> <td>   -0.638</td> <td>   -0.491</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[448.0]</th>    <td>   -0.0203</td> <td>    0.030</td> <td>   -0.669</td> <td> 0.504</td> <td>   -0.080</td> <td>    0.039</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[449.0]</th>    <td>    0.4358</td> <td>    0.039</td> <td>   11.178</td> <td> 0.000</td> <td>    0.359</td> <td>    0.512</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[452.0]</th>    <td>   -0.0755</td> <td>    0.028</td> <td>   -2.736</td> <td> 0.006</td> <td>   -0.130</td> <td>   -0.021</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[453.0]</th>    <td>    0.2476</td> <td>    0.039</td> <td>    6.331</td> <td> 0.000</td> <td>    0.171</td> <td>    0.324</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[454.0]</th>    <td>    0.1811</td> <td>    0.052</td> <td>    3.477</td> <td> 0.001</td> <td>    0.079</td> <td>    0.283</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[456.0]</th>    <td>   -0.6219</td> <td>    0.027</td> <td>  -22.984</td> <td> 0.000</td> <td>   -0.675</td> <td>   -0.569</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[465.0]</th>    <td>    0.6563</td> <td>    0.025</td> <td>   25.916</td> <td> 0.000</td> <td>    0.607</td> <td>    0.706</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[466.0]</th>    <td>    0.5222</td> <td>    0.035</td> <td>   15.026</td> <td> 0.000</td> <td>    0.454</td> <td>    0.590</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[469.0]</th>    <td>   -0.5433</td> <td>    0.027</td> <td>  -20.489</td> <td> 0.000</td> <td>   -0.595</td> <td>   -0.491</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[470.0]</th>    <td>   -0.0121</td> <td>    0.025</td> <td>   -0.489</td> <td> 0.625</td> <td>   -0.061</td> <td>    0.036</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[473.0]</th>    <td>   -0.0066</td> <td>    0.032</td> <td>   -0.207</td> <td> 0.836</td> <td>   -0.069</td> <td>    0.056</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[474.0]</th>    <td>   -0.0782</td> <td>    0.032</td> <td>   -2.454</td> <td> 0.014</td> <td>   -0.141</td> <td>   -0.016</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[475.0]</th>    <td>   -0.8366</td> <td>    0.033</td> <td>  -25.592</td> <td> 0.000</td> <td>   -0.901</td> <td>   -0.773</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[477.0]</th>    <td>   -0.2993</td> <td>    0.033</td> <td>   -9.205</td> <td> 0.000</td> <td>   -0.363</td> <td>   -0.236</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[479.0]</th>    <td>    0.3088</td> <td>    0.039</td> <td>    7.841</td> <td> 0.000</td> <td>    0.232</td> <td>    0.386</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[489.0]</th>    <td>    1.3348</td> <td>    0.031</td> <td>   43.235</td> <td> 0.000</td> <td>    1.274</td> <td>    1.395</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[493.0]</th>    <td>   -0.0813</td> <td>    0.018</td> <td>   -4.531</td> <td> 0.000</td> <td>   -0.116</td> <td>   -0.046</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[494.0]</th>    <td>    0.1038</td> <td>    0.046</td> <td>    2.240</td> <td> 0.025</td> <td>    0.013</td> <td>    0.195</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[496.0]</th>    <td>    1.0978</td> <td>    0.030</td> <td>   35.995</td> <td> 0.000</td> <td>    1.038</td> <td>    1.158</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[497.0]</th>    <td>   -0.2225</td> <td>    0.027</td> <td>   -8.366</td> <td> 0.000</td> <td>   -0.275</td> <td>   -0.170</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[499.0]</th>    <td>   -0.0460</td> <td>    0.049</td> <td>   -0.931</td> <td> 0.352</td> <td>   -0.143</td> <td>    0.051</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[503.0]</th>    <td>    0.2464</td> <td>    0.036</td> <td>    6.898</td> <td> 0.000</td> <td>    0.176</td> <td>    0.316</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[504.0]</th>    <td>   -0.1611</td> <td>    0.034</td> <td>   -4.769</td> <td> 0.000</td> <td>   -0.227</td> <td>   -0.095</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[507.0]</th>    <td>    0.8227</td> <td>    0.030</td> <td>   27.682</td> <td> 0.000</td> <td>    0.764</td> <td>    0.881</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[508.0]</th>    <td>   -0.3401</td> <td>    0.032</td> <td>  -10.632</td> <td> 0.000</td> <td>   -0.403</td> <td>   -0.277</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[511.0]</th>    <td>    0.1363</td> <td>    0.026</td> <td>    5.253</td> <td> 0.000</td> <td>    0.085</td> <td>    0.187</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[512.0]</th>    <td>    0.2864</td> <td>    0.033</td> <td>    8.608</td> <td> 0.000</td> <td>    0.221</td> <td>    0.352</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[514.0]</th>    <td>    0.1598</td> <td>    0.031</td> <td>    5.094</td> <td> 0.000</td> <td>    0.098</td> <td>    0.221</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[515.0]</th>    <td>    0.7828</td> <td>    0.041</td> <td>   18.931</td> <td> 0.000</td> <td>    0.702</td> <td>    0.864</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[517.0]</th>    <td>   -0.2875</td> <td>    0.033</td> <td>   -8.723</td> <td> 0.000</td> <td>   -0.352</td> <td>   -0.223</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[519.0]</th>    <td>    0.8525</td> <td>    0.034</td> <td>   25.200</td> <td> 0.000</td> <td>    0.786</td> <td>    0.919</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[521.0]</th>    <td>   -0.8118</td> <td>    0.031</td> <td>  -26.137</td> <td> 0.000</td> <td>   -0.873</td> <td>   -0.751</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[525.0]</th>    <td>    0.3071</td> <td>    0.028</td> <td>   10.935</td> <td> 0.000</td> <td>    0.252</td> <td>    0.362</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[526.0]</th>    <td>   -1.1562</td> <td>    0.035</td> <td>  -32.972</td> <td> 0.000</td> <td>   -1.225</td> <td>   -1.087</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[528.0]</th>    <td>    0.9126</td> <td>    0.031</td> <td>   29.343</td> <td> 0.000</td> <td>    0.852</td> <td>    0.974</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[529.0]</th>    <td>    0.0495</td> <td>    0.027</td> <td>    1.802</td> <td> 0.072</td> <td>   -0.004</td> <td>    0.103</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[534.0]</th>    <td>   -0.3113</td> <td>    0.029</td> <td>  -10.839</td> <td> 0.000</td> <td>   -0.368</td> <td>   -0.255</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[545.0]</th>    <td>    0.2825</td> <td>    0.030</td> <td>    9.308</td> <td> 0.000</td> <td>    0.223</td> <td>    0.342</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[546.0]</th>    <td>    0.0406</td> <td>    0.029</td> <td>    1.394</td> <td> 0.163</td> <td>   -0.016</td> <td>    0.098</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[549.0]</th>    <td>   -0.1461</td> <td>    0.030</td> <td>   -4.913</td> <td> 0.000</td> <td>   -0.204</td> <td>   -0.088</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[550.0]</th>    <td>   -0.0206</td> <td>    0.029</td> <td>   -0.722</td> <td> 0.470</td> <td>   -0.077</td> <td>    0.035</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[560.0]</th>    <td>    0.1188</td> <td>    0.026</td> <td>    4.655</td> <td> 0.000</td> <td>    0.069</td> <td>    0.169</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[564.0]</th>    <td>    1.6285</td> <td>    0.066</td> <td>   24.600</td> <td> 0.000</td> <td>    1.499</td> <td>    1.758</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[566.0]</th>    <td>    0.7114</td> <td>    0.049</td> <td>   14.445</td> <td> 0.000</td> <td>    0.615</td> <td>    0.808</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[567.0]</th>    <td>   -0.1149</td> <td>    0.036</td> <td>   -3.170</td> <td> 0.002</td> <td>   -0.186</td> <td>   -0.044</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[571.0]</th>    <td>    0.1440</td> <td>    0.014</td> <td>   10.158</td> <td> 0.000</td> <td>    0.116</td> <td>    0.172</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[574.0]</th>    <td>    0.0190</td> <td>    0.026</td> <td>    0.735</td> <td> 0.463</td> <td>   -0.032</td> <td>    0.070</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[575.0]</th>    <td>   -0.1625</td> <td>    0.025</td> <td>   -6.557</td> <td> 0.000</td> <td>   -0.211</td> <td>   -0.114</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[578.0]</th>    <td>    0.9040</td> <td>    0.031</td> <td>   29.527</td> <td> 0.000</td> <td>    0.844</td> <td>    0.964</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[580.0]</th>    <td>   -0.7299</td> <td>    0.034</td> <td>  -21.658</td> <td> 0.000</td> <td>   -0.796</td> <td>   -0.664</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[583.0]</th>    <td>    0.7934</td> <td>    0.033</td> <td>   23.873</td> <td> 0.000</td> <td>    0.728</td> <td>    0.859</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[584.0]</th>    <td>   -0.0663</td> <td>    0.016</td> <td>   -4.123</td> <td> 0.000</td> <td>   -0.098</td> <td>   -0.035</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[587.0]</th>    <td>   -0.6982</td> <td>    0.032</td> <td>  -21.947</td> <td> 0.000</td> <td>   -0.761</td> <td>   -0.636</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[589.0]</th>    <td>   -0.8384</td> <td>    0.032</td> <td>  -26.182</td> <td> 0.000</td> <td>   -0.901</td> <td>   -0.776</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[590.0]</th>    <td>    0.7272</td> <td>    0.033</td> <td>   22.370</td> <td> 0.000</td> <td>    0.663</td> <td>    0.791</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[593.0]</th>    <td>   -0.3975</td> <td>    0.031</td> <td>  -12.924</td> <td> 0.000</td> <td>   -0.458</td> <td>   -0.337</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[594.0]</th>    <td>   -0.5958</td> <td>    0.030</td> <td>  -19.988</td> <td> 0.000</td> <td>   -0.654</td> <td>   -0.537</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[596.0]</th>    <td>   -0.1780</td> <td>    0.025</td> <td>   -7.116</td> <td> 0.000</td> <td>   -0.227</td> <td>   -0.129</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[597.0]</th>    <td>   -0.5538</td> <td>    0.040</td> <td>  -13.699</td> <td> 0.000</td> <td>   -0.633</td> <td>   -0.475</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[598.0]</th>    <td>   -0.0187</td> <td>    0.024</td> <td>   -0.768</td> <td> 0.443</td> <td>   -0.067</td> <td>    0.029</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[602.0]</th>    <td>   -0.6033</td> <td>    0.028</td> <td>  -21.483</td> <td> 0.000</td> <td>   -0.658</td> <td>   -0.548</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[604.0]</th>    <td>    0.1874</td> <td>    0.033</td> <td>    5.630</td> <td> 0.000</td> <td>    0.122</td> <td>    0.253</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[608.0]</th>    <td>    0.5710</td> <td>    0.045</td> <td>   12.831</td> <td> 0.000</td> <td>    0.484</td> <td>    0.658</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[611.0]</th>    <td>   -0.2688</td> <td>    0.016</td> <td>  -16.592</td> <td> 0.000</td> <td>   -0.300</td> <td>   -0.237</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[613.0]</th>    <td>    0.1441</td> <td>    0.039</td> <td>    3.662</td> <td> 0.000</td> <td>    0.067</td> <td>    0.221</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[617.0]</th>    <td>   -0.1876</td> <td>    0.025</td> <td>   -7.641</td> <td> 0.000</td> <td>   -0.236</td> <td>   -0.139</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[620.0]</th>    <td>   -0.3154</td> <td>    0.028</td> <td>  -11.393</td> <td> 0.000</td> <td>   -0.370</td> <td>   -0.261</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[621.0]</th>    <td>   -0.8531</td> <td>    0.034</td> <td>  -24.896</td> <td> 0.000</td> <td>   -0.920</td> <td>   -0.786</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[622.0]</th>    <td>    1.1565</td> <td>    0.058</td> <td>   20.105</td> <td> 0.000</td> <td>    1.044</td> <td>    1.269</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[623.0]</th>    <td>   -0.7363</td> <td>    0.018</td> <td>  -40.484</td> <td> 0.000</td> <td>   -0.772</td> <td>   -0.701</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[625.0]</th>    <td>    0.8382</td> <td>    0.041</td> <td>   20.633</td> <td> 0.000</td> <td>    0.759</td> <td>    0.918</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[630.0]</th>    <td>   -0.4618</td> <td>    0.033</td> <td>  -14.127</td> <td> 0.000</td> <td>   -0.526</td> <td>   -0.398</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[632.0]</th>    <td>   -0.5798</td> <td>    0.022</td> <td>  -26.172</td> <td> 0.000</td> <td>   -0.623</td> <td>   -0.536</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[635.0]</th>    <td>   -0.0303</td> <td>    0.027</td> <td>   -1.113</td> <td> 0.266</td> <td>   -0.084</td> <td>    0.023</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[638.0]</th>    <td>   -0.1179</td> <td>    0.032</td> <td>   -3.711</td> <td> 0.000</td> <td>   -0.180</td> <td>   -0.056</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[642.0]</th>    <td>   -0.2363</td> <td>    0.025</td> <td>   -9.363</td> <td> 0.000</td> <td>   -0.286</td> <td>   -0.187</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[643.0]</th>    <td>    0.5519</td> <td>    0.032</td> <td>   17.032</td> <td> 0.000</td> <td>    0.488</td> <td>    0.615</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[645.0]</th>    <td>    0.3910</td> <td>    0.026</td> <td>   14.860</td> <td> 0.000</td> <td>    0.339</td> <td>    0.443</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[646.0]</th>    <td>   -0.2654</td> <td>    0.027</td> <td>   -9.789</td> <td> 0.000</td> <td>   -0.319</td> <td>   -0.212</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[649.0]</th>    <td>   -0.1261</td> <td>    0.016</td> <td>   -8.122</td> <td> 0.000</td> <td>   -0.157</td> <td>   -0.096</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[650.0]</th>    <td>    0.2105</td> <td>    0.036</td> <td>    5.864</td> <td> 0.000</td> <td>    0.140</td> <td>    0.281</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[651.0]</th>    <td>   -0.1754</td> <td>    0.034</td> <td>   -5.090</td> <td> 0.000</td> <td>   -0.243</td> <td>   -0.108</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[653.0]</th>    <td>   -0.4590</td> <td>    0.030</td> <td>  -15.314</td> <td> 0.000</td> <td>   -0.518</td> <td>   -0.400</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[655.0]</th>    <td>   -1.0367</td> <td>    0.037</td> <td>  -28.049</td> <td> 0.000</td> <td>   -1.109</td> <td>   -0.964</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[656.0]</th>    <td>    0.0615</td> <td>    0.033</td> <td>    1.852</td> <td> 0.064</td> <td>   -0.004</td> <td>    0.127</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[658.0]</th>    <td>    0.7501</td> <td>    0.039</td> <td>   19.088</td> <td> 0.000</td> <td>    0.673</td> <td>    0.827</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[659.0]</th>    <td>    0.5080</td> <td>    0.035</td> <td>   14.541</td> <td> 0.000</td> <td>    0.440</td> <td>    0.577</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[661.0]</th>    <td>    0.1404</td> <td>    0.033</td> <td>    4.302</td> <td> 0.000</td> <td>    0.076</td> <td>    0.204</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[663.0]</th>    <td>   -0.0345</td> <td>    0.040</td> <td>   -0.870</td> <td> 0.384</td> <td>   -0.112</td> <td>    0.043</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[665.0]</th>    <td>    0.3982</td> <td>    0.034</td> <td>   11.860</td> <td> 0.000</td> <td>    0.332</td> <td>    0.464</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[668.0]</th>    <td>    1.2572</td> <td>    0.036</td> <td>   34.803</td> <td> 0.000</td> <td>    1.186</td> <td>    1.328</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[669.0]</th>    <td>    1.2107</td> <td>    0.049</td> <td>   24.571</td> <td> 0.000</td> <td>    1.114</td> <td>    1.307</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[673.0]</th>    <td>   -0.0776</td> <td>    0.033</td> <td>   -2.376</td> <td> 0.017</td> <td>   -0.142</td> <td>   -0.014</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[676.0]</th>    <td>    0.2001</td> <td>    0.041</td> <td>    4.878</td> <td> 0.000</td> <td>    0.120</td> <td>    0.281</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[682.0]</th>    <td>   -0.6388</td> <td>    0.033</td> <td>  -19.144</td> <td> 0.000</td> <td>   -0.704</td> <td>   -0.573</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[683.0]</th>    <td>   -0.3823</td> <td>    0.050</td> <td>   -7.640</td> <td> 0.000</td> <td>   -0.480</td> <td>   -0.284</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[684.0]</th>    <td>    0.9383</td> <td>    0.030</td> <td>   31.085</td> <td> 0.000</td> <td>    0.879</td> <td>    0.997</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>C(id)[690.0]</th>    <td>   -0.5482</td> <td>    0.026</td> <td>  -20.984</td> <td> 0.000</td> <td>   -0.599</td> <td>   -0.497</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>age</th>             <td>    0.2818</td> <td>    0.008</td> <td>   33.614</td> <td> 0.000</td> <td>    0.265</td> <td>    0.298</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asq</th>             <td>   -0.0037</td> <td>    0.000</td> <td>  -33.942</td> <td> 0.000</td> <td>   -0.004</td> <td>   -0.004</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>bmi</th>             <td>   -0.0177</td> <td>    0.001</td> <td>  -30.110</td> <td> 0.000</td> <td>   -0.019</td> <td>   -0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hispanic</th>        <td>    0.0707</td> <td>    0.010</td> <td>    6.821</td> <td> 0.000</td> <td>    0.050</td> <td>    0.091</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black</th>           <td>    0.3222</td> <td>    0.016</td> <td>   20.587</td> <td> 0.000</td> <td>    0.292</td> <td>    0.353</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>other</th>           <td>   -0.0490</td> <td>    0.008</td> <td>   -5.898</td> <td> 0.000</td> <td>   -0.065</td> <td>   -0.033</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asian</th>           <td>    0.0777</td> <td>    0.015</td> <td>    5.213</td> <td> 0.000</td> <td>    0.048</td> <td>    0.107</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>schooling</th>       <td>    0.1853</td> <td>    0.005</td> <td>   35.956</td> <td> 0.000</td> <td>    0.175</td> <td>    0.195</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>cohab</th>           <td>   -0.0562</td> <td>    0.006</td> <td>   -9.063</td> <td> 0.000</td> <td>   -0.068</td> <td>   -0.044</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>married</th>         <td>   -0.0561</td> <td>    0.007</td> <td>   -7.575</td> <td> 0.000</td> <td>   -0.071</td> <td>   -0.042</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>divorced</th>        <td>   -0.1357</td> <td>    0.006</td> <td>  -21.623</td> <td> 0.000</td> <td>   -0.148</td> <td>   -0.123</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>separated</th>       <td>    0.1397</td> <td>    0.005</td> <td>   26.200</td> <td> 0.000</td> <td>    0.129</td> <td>    0.150</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>age_cl</th>          <td>    0.0023</td> <td>    0.008</td> <td>    0.292</td> <td> 0.770</td> <td>   -0.013</td> <td>    0.018</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>unsafe</th>          <td>    0.0510</td> <td>    0.033</td> <td>    1.548</td> <td> 0.122</td> <td>   -0.014</td> <td>    0.116</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>llength</th>         <td>   -0.4345</td> <td>    0.028</td> <td>  -15.323</td> <td> 0.000</td> <td>   -0.490</td> <td>   -0.379</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>reg</th>             <td>   -0.0373</td> <td>    0.022</td> <td>   -1.707</td> <td> 0.088</td> <td>   -0.080</td> <td>    0.006</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asq_cl</th>          <td> -1.47e-05</td> <td>  8.8e-05</td> <td>   -0.167</td> <td> 0.867</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>appearance_cl</th>   <td>    0.0056</td> <td>    0.007</td> <td>    0.833</td> <td> 0.405</td> <td>   -0.008</td> <td>    0.019</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>provider_second</th> <td>    0.1131</td> <td>    0.070</td> <td>    1.607</td> <td> 0.108</td> <td>   -0.025</td> <td>    0.251</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>asian_cl</th>        <td>   -0.0099</td> <td>    0.039</td> <td>   -0.250</td> <td> 0.802</td> <td>   -0.087</td> <td>    0.067</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>black_cl</th>        <td>    0.0265</td> <td>    0.049</td> <td>    0.537</td> <td> 0.591</td> <td>   -0.070</td> <td>    0.123</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hispanic_cl</th>     <td>   -0.0621</td> <td>    0.060</td> <td>   -1.030</td> <td> 0.303</td> <td>   -0.180</td> <td>    0.056</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>othrace_cl</th>      <td>    0.1422</td> <td>    0.057</td> <td>    2.474</td> <td> 0.013</td> <td>    0.030</td> <td>    0.255</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>hot</th>             <td>    0.0525</td> <td>    0.032</td> <td>    1.660</td> <td> 0.097</td> <td>   -0.009</td> <td>    0.114</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>massage_cl</th>      <td>   -0.0010</td> <td>    0.032</td> <td>   -0.032</td> <td> 0.975</td> <td>   -0.065</td> <td>    0.062</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>79.779</td> <th>  Durbin-Watson:     </th> <td>   2.528</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 361.895</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.172</td> <th>  Prob(JB):          </th> <td>2.60e-79</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 5.886</td> <th>  Cond. No.          </th> <td>9.19e+20</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 9.08e-33. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                    lnw   R-squared:                       0.832\n",
+       "Model:                            OLS   Adj. R-squared:                  0.773\n",
+       "Method:                 Least Squares   F-statistic:                       nan\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):                nan\n",
+       "Time:                        13:32:50   Log-Likelihood:                 162.25\n",
+       "No. Observations:                1028   AIC:                             215.5\n",
+       "Df Residuals:                     758   BIC:                             1548.\n",
+       "Df Model:                         269                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "===================================================================================\n",
+       "                      coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "-----------------------------------------------------------------------------------\n",
+       "C(id)[6.0]         -0.4465      0.035    -12.683      0.000      -0.515      -0.377\n",
+       "C(id)[8.0]          0.3310      0.027     12.240      0.000       0.278       0.384\n",
+       "C(id)[10.0]         1.0513      0.040     26.539      0.000       0.974       1.129\n",
+       "C(id)[11.0]        -0.5627      0.026    -21.608      0.000      -0.614      -0.512\n",
+       "C(id)[18.0]         0.5518      0.034     16.025      0.000       0.484       0.619\n",
+       "C(id)[23.0]        -0.0312      0.038     -0.827      0.408      -0.105       0.043\n",
+       "C(id)[25.0]        -0.1525      0.035     -4.403      0.000      -0.220      -0.085\n",
+       "C(id)[29.0]         1.6517      0.068     24.285      0.000       1.518       1.785\n",
+       "C(id)[31.0]        -0.0586      0.020     -2.953      0.003      -0.098      -0.020\n",
+       "C(id)[33.0]         0.3145      0.021     14.914      0.000       0.273       0.356\n",
+       "C(id)[34.0]        -0.4864      0.030    -16.439      0.000      -0.544      -0.428\n",
+       "C(id)[35.0]         0.8514      0.040     21.275      0.000       0.773       0.930\n",
+       "C(id)[42.0]         0.8090      0.040     20.387      0.000       0.731       0.887\n",
+       "C(id)[44.0]        -0.3047      0.035     -8.649      0.000      -0.374      -0.236\n",
+       "C(id)[48.0]        -0.3631      0.030    -12.024      0.000      -0.422      -0.304\n",
+       "C(id)[53.0]         0.3511      0.031     11.184      0.000       0.290       0.413\n",
+       "C(id)[54.0]         0.6716      0.036     18.840      0.000       0.602       0.741\n",
+       "C(id)[56.0]        -0.4843      0.024    -20.443      0.000      -0.531      -0.438\n",
+       "C(id)[59.0]        -0.3581      0.025    -14.086      0.000      -0.408      -0.308\n",
+       "C(id)[61.0]        -0.4522      0.057     -7.929      0.000      -0.564      -0.340\n",
+       "C(id)[62.0]        -0.2078      0.033     -6.232      0.000      -0.273      -0.142\n",
+       "C(id)[64.0]        -0.3638      0.038     -9.589      0.000      -0.438      -0.289\n",
+       "C(id)[68.0]         0.2320      0.028      8.163      0.000       0.176       0.288\n",
+       "C(id)[72.0]         0.0063      0.033      0.191      0.849      -0.058       0.070\n",
+       "C(id)[73.0]        -0.1538      0.030     -5.065      0.000      -0.213      -0.094\n",
+       "C(id)[77.0]        -0.4928      0.037    -13.168      0.000      -0.566      -0.419\n",
+       "C(id)[78.0]        -0.5062      0.034    -14.938      0.000      -0.573      -0.440\n",
+       "C(id)[79.0]         0.9076      0.044     20.816      0.000       0.822       0.993\n",
+       "C(id)[83.0]        -0.4709      0.015    -30.601      0.000      -0.501      -0.441\n",
+       "C(id)[84.0]        -0.8525      0.037    -22.855      0.000      -0.926      -0.779\n",
+       "C(id)[86.0]         0.7815      0.036     21.836      0.000       0.711       0.852\n",
+       "C(id)[88.0]         0.5577      0.031     17.889      0.000       0.497       0.619\n",
+       "C(id)[91.0]        -0.7363      0.043    -16.985      0.000      -0.821      -0.651\n",
+       "C(id)[93.0]         0.2723      0.029      9.354      0.000       0.215       0.329\n",
+       "C(id)[94.0]         0.5675      0.050     11.372      0.000       0.470       0.665\n",
+       "C(id)[96.0]         0.5108      0.026     19.505      0.000       0.459       0.562\n",
+       "C(id)[100.0]       -0.2375      0.041     -5.815      0.000      -0.318      -0.157\n",
+       "C(id)[105.0]        0.6224      0.031     19.909      0.000       0.561       0.684\n",
+       "C(id)[107.0]       -0.8367      0.041    -20.566      0.000      -0.916      -0.757\n",
+       "C(id)[108.0]       -0.4168      0.050     -8.403      0.000      -0.514      -0.320\n",
+       "C(id)[113.0]        0.0119      0.020      0.582      0.561      -0.028       0.052\n",
+       "C(id)[114.0]       -0.1436      0.028     -5.102      0.000      -0.199      -0.088\n",
+       "C(id)[115.0]        0.2745      0.042      6.474      0.000       0.191       0.358\n",
+       "C(id)[120.0]       -0.0830      0.025     -3.334      0.001      -0.132      -0.034\n",
+       "C(id)[121.0]        0.0400      0.042      0.940      0.347      -0.043       0.123\n",
+       "C(id)[122.0]        0.2025      0.030      6.683      0.000       0.143       0.262\n",
+       "C(id)[123.0]        0.1032      0.012      8.494      0.000       0.079       0.127\n",
+       "C(id)[128.0]        0.1970      0.027      7.224      0.000       0.144       0.250\n",
+       "C(id)[130.0]       -0.8677      0.034    -25.749      0.000      -0.934      -0.802\n",
+       "C(id)[131.0]       -1.0605      0.040    -26.442      0.000      -1.139      -0.982\n",
+       "C(id)[134.0]       -0.3681      0.031    -11.754      0.000      -0.429      -0.307\n",
+       "C(id)[137.0]       -1.1538      0.043    -26.747      0.000      -1.238      -1.069\n",
+       "C(id)[138.0]        0.8507      0.027     31.608      0.000       0.798       0.903\n",
+       "C(id)[140.0]       -0.7737      0.040    -19.503      0.000      -0.851      -0.696\n",
+       "C(id)[145.0]       -0.5292      0.033    -16.207      0.000      -0.593      -0.465\n",
+       "C(id)[148.0]        0.7798      0.036     21.739      0.000       0.709       0.850\n",
+       "C(id)[149.0]        0.4930      0.033     15.156      0.000       0.429       0.557\n",
+       "C(id)[150.0]        1.3854      0.048     29.121      0.000       1.292       1.479\n",
+       "C(id)[151.0]       -0.1263      0.030     -4.144      0.000      -0.186      -0.067\n",
+       "C(id)[153.0]       -0.3541      0.034    -10.319      0.000      -0.421      -0.287\n",
+       "C(id)[156.0]        0.1443      0.035      4.113      0.000       0.076       0.213\n",
+       "C(id)[158.0]       -0.5936      0.022    -26.833      0.000      -0.637      -0.550\n",
+       "C(id)[159.0]       -0.1781      0.039     -4.525      0.000      -0.255      -0.101\n",
+       "C(id)[164.0]        0.1872      0.031      6.136      0.000       0.127       0.247\n",
+       "C(id)[165.0]        0.7737      0.029     26.643      0.000       0.717       0.831\n",
+       "C(id)[166.0]       -0.2982      0.025    -12.043      0.000      -0.347      -0.250\n",
+       "C(id)[168.0]       -0.9939      0.046    -21.529      0.000      -1.084      -0.903\n",
+       "C(id)[170.0]        0.4943      0.025     19.974      0.000       0.446       0.543\n",
+       "C(id)[173.0]        0.7285      0.027     27.433      0.000       0.676       0.781\n",
+       "C(id)[176.0]       -0.3965      0.027    -14.515      0.000      -0.450      -0.343\n",
+       "C(id)[177.0]        0.0037      0.027      0.136      0.892      -0.049       0.057\n",
+       "C(id)[179.0]       -0.7709      0.035    -22.034      0.000      -0.840      -0.702\n",
+       "C(id)[180.0]       -0.2945      0.033     -8.922      0.000      -0.359      -0.230\n",
+       "C(id)[181.0]        0.2046      0.024      8.695      0.000       0.159       0.251\n",
+       "C(id)[183.0]       -0.0737      0.026     -2.785      0.005      -0.126      -0.022\n",
+       "C(id)[185.0]        0.7862      0.041     19.218      0.000       0.706       0.866\n",
+       "C(id)[186.0]       -0.3200      0.033     -9.666      0.000      -0.385      -0.255\n",
+       "C(id)[188.0]        0.4196      0.041     10.346      0.000       0.340       0.499\n",
+       "C(id)[189.0]       -0.3912      0.055     -7.150      0.000      -0.498      -0.284\n",
+       "C(id)[191.0]        0.0024      0.036      0.067      0.947      -0.069       0.074\n",
+       "C(id)[193.0]       -0.7278      0.042    -17.355      0.000      -0.810      -0.646\n",
+       "C(id)[195.0]        0.4724      0.031     15.059      0.000       0.411       0.534\n",
+       "C(id)[199.0]       -0.8182      0.022    -37.782      0.000      -0.861      -0.776\n",
+       "C(id)[200.0]        0.1035      0.028      3.701      0.000       0.049       0.158\n",
+       "C(id)[202.0]        0.3100      0.032      9.700      0.000       0.247       0.373\n",
+       "C(id)[206.0]        0.2463      0.028      8.652      0.000       0.190       0.302\n",
+       "C(id)[215.0]       -0.3703      0.026    -14.404      0.000      -0.421      -0.320\n",
+       "C(id)[220.0]       -0.0165      0.028     -0.581      0.561      -0.072       0.039\n",
+       "C(id)[223.0]        0.3535      0.041      8.533      0.000       0.272       0.435\n",
+       "C(id)[225.0]        0.1590      0.019      8.565      0.000       0.123       0.195\n",
+       "C(id)[227.0]        0.1842      0.034      5.361      0.000       0.117       0.252\n",
+       "C(id)[229.0]       -0.3853      0.023    -17.076      0.000      -0.429      -0.341\n",
+       "C(id)[230.0]        0.3953      0.032     12.419      0.000       0.333       0.458\n",
+       "C(id)[231.0]        0.0641      0.034      1.874      0.061      -0.003       0.131\n",
+       "C(id)[238.0]        0.1458      0.050      2.906      0.004       0.047       0.244\n",
+       "C(id)[239.0]       -0.4659      0.045    -10.334      0.000      -0.554      -0.378\n",
+       "C(id)[246.0]       -0.0273      0.023     -1.199      0.231      -0.072       0.017\n",
+       "C(id)[253.0]       -0.7272      0.039    -18.829      0.000      -0.803      -0.651\n",
+       "C(id)[257.0]       -0.1557      0.027     -5.671      0.000      -0.210      -0.102\n",
+       "C(id)[271.0]       -0.2892      0.021    -14.042      0.000      -0.330      -0.249\n",
+       "C(id)[274.0]        0.2065      0.038      5.448      0.000       0.132       0.281\n",
+       "C(id)[279.0]       -0.5520      0.037    -14.896      0.000      -0.625      -0.479\n",
+       "C(id)[280.0]        0.4347      0.032     13.642      0.000       0.372       0.497\n",
+       "C(id)[281.0]        0.4606      0.040     11.514      0.000       0.382       0.539\n",
+       "C(id)[282.0]       -0.0460      0.023     -2.031      0.042      -0.090      -0.002\n",
+       "C(id)[283.0]       -0.1643      0.024     -6.935      0.000      -0.211      -0.118\n",
+       "C(id)[292.0]        0.3708      0.040      9.307      0.000       0.293       0.449\n",
+       "C(id)[295.0]        1.4466      0.028     51.492      0.000       1.391       1.502\n",
+       "C(id)[298.0]        1.3459      0.042     32.059      0.000       1.264       1.428\n",
+       "C(id)[299.0]        0.2884      0.033      8.680      0.000       0.223       0.354\n",
+       "C(id)[303.0]       -0.6055      0.029    -20.985      0.000      -0.662      -0.549\n",
+       "C(id)[304.0]        0.0778      0.029      2.642      0.008       0.020       0.136\n",
+       "C(id)[306.0]        0.5709      0.031     18.691      0.000       0.511       0.631\n",
+       "C(id)[308.0]        0.3815      0.028     13.443      0.000       0.326       0.437\n",
+       "C(id)[309.0]        0.0374      0.029      1.301      0.193      -0.019       0.094\n",
+       "C(id)[310.0]        1.3725      0.039     34.955      0.000       1.296       1.449\n",
+       "C(id)[314.0]       -0.3016      0.031     -9.730      0.000      -0.362      -0.241\n",
+       "C(id)[315.0]       -0.1683      0.033     -5.028      0.000      -0.234      -0.103\n",
+       "C(id)[334.0]       -0.1043      0.029     -3.542      0.000      -0.162      -0.047\n",
+       "C(id)[335.0]        0.3782      0.021     18.014      0.000       0.337       0.419\n",
+       "C(id)[337.0]       -0.9128      0.029    -31.659      0.000      -0.969      -0.856\n",
+       "C(id)[339.0]       -0.3775      0.037    -10.266      0.000      -0.450      -0.305\n",
+       "C(id)[340.0]        0.6570      0.043     15.280      0.000       0.573       0.741\n",
+       "C(id)[343.0]       -0.1178      0.028     -4.168      0.000      -0.173      -0.062\n",
+       "C(id)[345.0]        1.4559      0.029     50.267      0.000       1.399       1.513\n",
+       "C(id)[351.0]       -0.3798      0.042     -8.998      0.000      -0.463      -0.297\n",
+       "C(id)[357.0]        0.1202      0.035      3.429      0.001       0.051       0.189\n",
+       "C(id)[358.0]        0.3197      0.032      9.853      0.000       0.256       0.383\n",
+       "C(id)[361.0]       -0.2092      0.042     -4.970      0.000      -0.292      -0.127\n",
+       "C(id)[364.0]       -0.2560      0.055     -4.632      0.000      -0.364      -0.148\n",
+       "C(id)[365.0]       -0.3782      0.026    -14.416      0.000      -0.430      -0.327\n",
+       "C(id)[368.0]       -0.3443      0.037     -9.335      0.000      -0.417      -0.272\n",
+       "C(id)[370.0]       -0.0859      0.022     -3.933      0.000      -0.129      -0.043\n",
+       "C(id)[371.0]        0.7179      0.041     17.335      0.000       0.637       0.799\n",
+       "C(id)[373.0]        0.2087      0.023      8.924      0.000       0.163       0.255\n",
+       "C(id)[376.0]       -0.5252      0.038    -13.673      0.000      -0.600      -0.450\n",
+       "C(id)[378.0]        0.2573      0.033      7.689      0.000       0.192       0.323\n",
+       "C(id)[381.0]        1.0624      0.026     40.503      0.000       1.011       1.114\n",
+       "C(id)[382.0]       -0.8283      0.042    -19.778      0.000      -0.910      -0.746\n",
+       "C(id)[388.0]       -0.7960      0.038    -21.018      0.000      -0.870      -0.722\n",
+       "C(id)[393.0]       -0.5925      0.035    -16.706      0.000      -0.662      -0.523\n",
+       "C(id)[394.0]       -0.5698      0.029    -19.550      0.000      -0.627      -0.513\n",
+       "C(id)[395.0]       -0.0093      0.038     -0.244      0.807      -0.084       0.065\n",
+       "C(id)[396.0]       -0.9476      0.042    -22.397      0.000      -1.031      -0.865\n",
+       "C(id)[397.0]        0.0325      0.027      1.227      0.220      -0.019       0.085\n",
+       "C(id)[403.0]       -0.1233      0.032     -3.848      0.000      -0.186      -0.060\n",
+       "C(id)[404.0]        0.1958      0.029      6.853      0.000       0.140       0.252\n",
+       "C(id)[410.0]       -0.3268      0.011    -29.689      0.000      -0.348      -0.305\n",
+       "C(id)[416.0]        0.0475      0.031      1.518      0.129      -0.014       0.109\n",
+       "C(id)[421.0]        0.6134      0.029     21.068      0.000       0.556       0.670\n",
+       "C(id)[422.0]        0.3976      0.043      9.272      0.000       0.314       0.482\n",
+       "C(id)[424.0]       -0.7169      0.028    -25.183      0.000      -0.773      -0.661\n",
+       "C(id)[427.0]       -0.7037      0.030    -23.804      0.000      -0.762      -0.646\n",
+       "C(id)[432.0]       -0.9292      0.033    -28.090      0.000      -0.994      -0.864\n",
+       "C(id)[433.0]        0.1764      0.042      4.169      0.000       0.093       0.259\n",
+       "C(id)[440.0]        0.2746      0.042      6.583      0.000       0.193       0.356\n",
+       "C(id)[441.0]        0.2780      0.028      9.806      0.000       0.222       0.334\n",
+       "C(id)[443.0]       -0.3816      0.017    -22.708      0.000      -0.415      -0.349\n",
+       "C(id)[445.0]       -0.5643      0.037    -15.090      0.000      -0.638      -0.491\n",
+       "C(id)[448.0]       -0.0203      0.030     -0.669      0.504      -0.080       0.039\n",
+       "C(id)[449.0]        0.4358      0.039     11.178      0.000       0.359       0.512\n",
+       "C(id)[452.0]       -0.0755      0.028     -2.736      0.006      -0.130      -0.021\n",
+       "C(id)[453.0]        0.2476      0.039      6.331      0.000       0.171       0.324\n",
+       "C(id)[454.0]        0.1811      0.052      3.477      0.001       0.079       0.283\n",
+       "C(id)[456.0]       -0.6219      0.027    -22.984      0.000      -0.675      -0.569\n",
+       "C(id)[465.0]        0.6563      0.025     25.916      0.000       0.607       0.706\n",
+       "C(id)[466.0]        0.5222      0.035     15.026      0.000       0.454       0.590\n",
+       "C(id)[469.0]       -0.5433      0.027    -20.489      0.000      -0.595      -0.491\n",
+       "C(id)[470.0]       -0.0121      0.025     -0.489      0.625      -0.061       0.036\n",
+       "C(id)[473.0]       -0.0066      0.032     -0.207      0.836      -0.069       0.056\n",
+       "C(id)[474.0]       -0.0782      0.032     -2.454      0.014      -0.141      -0.016\n",
+       "C(id)[475.0]       -0.8366      0.033    -25.592      0.000      -0.901      -0.773\n",
+       "C(id)[477.0]       -0.2993      0.033     -9.205      0.000      -0.363      -0.236\n",
+       "C(id)[479.0]        0.3088      0.039      7.841      0.000       0.232       0.386\n",
+       "C(id)[489.0]        1.3348      0.031     43.235      0.000       1.274       1.395\n",
+       "C(id)[493.0]       -0.0813      0.018     -4.531      0.000      -0.116      -0.046\n",
+       "C(id)[494.0]        0.1038      0.046      2.240      0.025       0.013       0.195\n",
+       "C(id)[496.0]        1.0978      0.030     35.995      0.000       1.038       1.158\n",
+       "C(id)[497.0]       -0.2225      0.027     -8.366      0.000      -0.275      -0.170\n",
+       "C(id)[499.0]       -0.0460      0.049     -0.931      0.352      -0.143       0.051\n",
+       "C(id)[503.0]        0.2464      0.036      6.898      0.000       0.176       0.316\n",
+       "C(id)[504.0]       -0.1611      0.034     -4.769      0.000      -0.227      -0.095\n",
+       "C(id)[507.0]        0.8227      0.030     27.682      0.000       0.764       0.881\n",
+       "C(id)[508.0]       -0.3401      0.032    -10.632      0.000      -0.403      -0.277\n",
+       "C(id)[511.0]        0.1363      0.026      5.253      0.000       0.085       0.187\n",
+       "C(id)[512.0]        0.2864      0.033      8.608      0.000       0.221       0.352\n",
+       "C(id)[514.0]        0.1598      0.031      5.094      0.000       0.098       0.221\n",
+       "C(id)[515.0]        0.7828      0.041     18.931      0.000       0.702       0.864\n",
+       "C(id)[517.0]       -0.2875      0.033     -8.723      0.000      -0.352      -0.223\n",
+       "C(id)[519.0]        0.8525      0.034     25.200      0.000       0.786       0.919\n",
+       "C(id)[521.0]       -0.8118      0.031    -26.137      0.000      -0.873      -0.751\n",
+       "C(id)[525.0]        0.3071      0.028     10.935      0.000       0.252       0.362\n",
+       "C(id)[526.0]       -1.1562      0.035    -32.972      0.000      -1.225      -1.087\n",
+       "C(id)[528.0]        0.9126      0.031     29.343      0.000       0.852       0.974\n",
+       "C(id)[529.0]        0.0495      0.027      1.802      0.072      -0.004       0.103\n",
+       "C(id)[534.0]       -0.3113      0.029    -10.839      0.000      -0.368      -0.255\n",
+       "C(id)[545.0]        0.2825      0.030      9.308      0.000       0.223       0.342\n",
+       "C(id)[546.0]        0.0406      0.029      1.394      0.163      -0.016       0.098\n",
+       "C(id)[549.0]       -0.1461      0.030     -4.913      0.000      -0.204      -0.088\n",
+       "C(id)[550.0]       -0.0206      0.029     -0.722      0.470      -0.077       0.035\n",
+       "C(id)[560.0]        0.1188      0.026      4.655      0.000       0.069       0.169\n",
+       "C(id)[564.0]        1.6285      0.066     24.600      0.000       1.499       1.758\n",
+       "C(id)[566.0]        0.7114      0.049     14.445      0.000       0.615       0.808\n",
+       "C(id)[567.0]       -0.1149      0.036     -3.170      0.002      -0.186      -0.044\n",
+       "C(id)[571.0]        0.1440      0.014     10.158      0.000       0.116       0.172\n",
+       "C(id)[574.0]        0.0190      0.026      0.735      0.463      -0.032       0.070\n",
+       "C(id)[575.0]       -0.1625      0.025     -6.557      0.000      -0.211      -0.114\n",
+       "C(id)[578.0]        0.9040      0.031     29.527      0.000       0.844       0.964\n",
+       "C(id)[580.0]       -0.7299      0.034    -21.658      0.000      -0.796      -0.664\n",
+       "C(id)[583.0]        0.7934      0.033     23.873      0.000       0.728       0.859\n",
+       "C(id)[584.0]       -0.0663      0.016     -4.123      0.000      -0.098      -0.035\n",
+       "C(id)[587.0]       -0.6982      0.032    -21.947      0.000      -0.761      -0.636\n",
+       "C(id)[589.0]       -0.8384      0.032    -26.182      0.000      -0.901      -0.776\n",
+       "C(id)[590.0]        0.7272      0.033     22.370      0.000       0.663       0.791\n",
+       "C(id)[593.0]       -0.3975      0.031    -12.924      0.000      -0.458      -0.337\n",
+       "C(id)[594.0]       -0.5958      0.030    -19.988      0.000      -0.654      -0.537\n",
+       "C(id)[596.0]       -0.1780      0.025     -7.116      0.000      -0.227      -0.129\n",
+       "C(id)[597.0]       -0.5538      0.040    -13.699      0.000      -0.633      -0.475\n",
+       "C(id)[598.0]       -0.0187      0.024     -0.768      0.443      -0.067       0.029\n",
+       "C(id)[602.0]       -0.6033      0.028    -21.483      0.000      -0.658      -0.548\n",
+       "C(id)[604.0]        0.1874      0.033      5.630      0.000       0.122       0.253\n",
+       "C(id)[608.0]        0.5710      0.045     12.831      0.000       0.484       0.658\n",
+       "C(id)[611.0]       -0.2688      0.016    -16.592      0.000      -0.300      -0.237\n",
+       "C(id)[613.0]        0.1441      0.039      3.662      0.000       0.067       0.221\n",
+       "C(id)[617.0]       -0.1876      0.025     -7.641      0.000      -0.236      -0.139\n",
+       "C(id)[620.0]       -0.3154      0.028    -11.393      0.000      -0.370      -0.261\n",
+       "C(id)[621.0]       -0.8531      0.034    -24.896      0.000      -0.920      -0.786\n",
+       "C(id)[622.0]        1.1565      0.058     20.105      0.000       1.044       1.269\n",
+       "C(id)[623.0]       -0.7363      0.018    -40.484      0.000      -0.772      -0.701\n",
+       "C(id)[625.0]        0.8382      0.041     20.633      0.000       0.759       0.918\n",
+       "C(id)[630.0]       -0.4618      0.033    -14.127      0.000      -0.526      -0.398\n",
+       "C(id)[632.0]       -0.5798      0.022    -26.172      0.000      -0.623      -0.536\n",
+       "C(id)[635.0]       -0.0303      0.027     -1.113      0.266      -0.084       0.023\n",
+       "C(id)[638.0]       -0.1179      0.032     -3.711      0.000      -0.180      -0.056\n",
+       "C(id)[642.0]       -0.2363      0.025     -9.363      0.000      -0.286      -0.187\n",
+       "C(id)[643.0]        0.5519      0.032     17.032      0.000       0.488       0.615\n",
+       "C(id)[645.0]        0.3910      0.026     14.860      0.000       0.339       0.443\n",
+       "C(id)[646.0]       -0.2654      0.027     -9.789      0.000      -0.319      -0.212\n",
+       "C(id)[649.0]       -0.1261      0.016     -8.122      0.000      -0.157      -0.096\n",
+       "C(id)[650.0]        0.2105      0.036      5.864      0.000       0.140       0.281\n",
+       "C(id)[651.0]       -0.1754      0.034     -5.090      0.000      -0.243      -0.108\n",
+       "C(id)[653.0]       -0.4590      0.030    -15.314      0.000      -0.518      -0.400\n",
+       "C(id)[655.0]       -1.0367      0.037    -28.049      0.000      -1.109      -0.964\n",
+       "C(id)[656.0]        0.0615      0.033      1.852      0.064      -0.004       0.127\n",
+       "C(id)[658.0]        0.7501      0.039     19.088      0.000       0.673       0.827\n",
+       "C(id)[659.0]        0.5080      0.035     14.541      0.000       0.440       0.577\n",
+       "C(id)[661.0]        0.1404      0.033      4.302      0.000       0.076       0.204\n",
+       "C(id)[663.0]       -0.0345      0.040     -0.870      0.384      -0.112       0.043\n",
+       "C(id)[665.0]        0.3982      0.034     11.860      0.000       0.332       0.464\n",
+       "C(id)[668.0]        1.2572      0.036     34.803      0.000       1.186       1.328\n",
+       "C(id)[669.0]        1.2107      0.049     24.571      0.000       1.114       1.307\n",
+       "C(id)[673.0]       -0.0776      0.033     -2.376      0.017      -0.142      -0.014\n",
+       "C(id)[676.0]        0.2001      0.041      4.878      0.000       0.120       0.281\n",
+       "C(id)[682.0]       -0.6388      0.033    -19.144      0.000      -0.704      -0.573\n",
+       "C(id)[683.0]       -0.3823      0.050     -7.640      0.000      -0.480      -0.284\n",
+       "C(id)[684.0]        0.9383      0.030     31.085      0.000       0.879       0.997\n",
+       "C(id)[690.0]       -0.5482      0.026    -20.984      0.000      -0.599      -0.497\n",
+       "age                 0.2818      0.008     33.614      0.000       0.265       0.298\n",
+       "asq                -0.0037      0.000    -33.942      0.000      -0.004      -0.004\n",
+       "bmi                -0.0177      0.001    -30.110      0.000      -0.019      -0.017\n",
+       "hispanic            0.0707      0.010      6.821      0.000       0.050       0.091\n",
+       "black               0.3222      0.016     20.587      0.000       0.292       0.353\n",
+       "other              -0.0490      0.008     -5.898      0.000      -0.065      -0.033\n",
+       "asian               0.0777      0.015      5.213      0.000       0.048       0.107\n",
+       "schooling           0.1853      0.005     35.956      0.000       0.175       0.195\n",
+       "cohab              -0.0562      0.006     -9.063      0.000      -0.068      -0.044\n",
+       "married            -0.0561      0.007     -7.575      0.000      -0.071      -0.042\n",
+       "divorced           -0.1357      0.006    -21.623      0.000      -0.148      -0.123\n",
+       "separated           0.1397      0.005     26.200      0.000       0.129       0.150\n",
+       "age_cl              0.0023      0.008      0.292      0.770      -0.013       0.018\n",
+       "unsafe              0.0510      0.033      1.548      0.122      -0.014       0.116\n",
+       "llength            -0.4345      0.028    -15.323      0.000      -0.490      -0.379\n",
+       "reg                -0.0373      0.022     -1.707      0.088      -0.080       0.006\n",
+       "asq_cl           -1.47e-05    8.8e-05     -0.167      0.867      -0.000       0.000\n",
+       "appearance_cl       0.0056      0.007      0.833      0.405      -0.008       0.019\n",
+       "provider_second     0.1131      0.070      1.607      0.108      -0.025       0.251\n",
+       "asian_cl           -0.0099      0.039     -0.250      0.802      -0.087       0.067\n",
+       "black_cl            0.0265      0.049      0.537      0.591      -0.070       0.123\n",
+       "hispanic_cl        -0.0621      0.060     -1.030      0.303      -0.180       0.056\n",
+       "othrace_cl          0.1422      0.057      2.474      0.013       0.030       0.255\n",
+       "hot                 0.0525      0.032      1.660      0.097      -0.009       0.114\n",
+       "massage_cl         -0.0010      0.032     -0.032      0.975      -0.065       0.062\n",
+       "==============================================================================\n",
+       "Omnibus:                       79.779   Durbin-Watson:                   2.528\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              361.895\n",
+       "Skew:                           0.172   Prob(JB):                     2.60e-79\n",
+       "Kurtosis:                       5.886   Cond. No.                     9.19e+20\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 9.08e-33. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "formula = \"\"\"lnw ~ -1 + C(id) + age + asq + bmi + hispanic + black + other + asian + schooling + \n",
+    "                      cohab + married + divorced + separated + \n",
+    "                      age_cl + unsafe + llength + reg + asq_cl + appearance_cl + \n",
+    "                      provider_second + asian_cl + black_cl + hispanic_cl + \n",
+    "                      othrace_cl + hot + massage_cl\"\"\"\n",
+    "\n",
+    "ols = sm.OLS.from_formula(formula, data=balanced_sasp).fit(cov_type='cluster', \n",
+    "                                                           cov_kwds={'groups': balanced_sasp['id']})\n",
+    "ols.summary()    "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Demean OLS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/statsmodels/base/model.py:1832: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 25, but rank is 13\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>       <td>demean_lnw</td>    <th>  R-squared:         </th> <td>   0.516</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.510</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   32.39</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>9.91e-47</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:32:50</td>     <th>  Log-Likelihood:    </th> <td>  162.25</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  1028</td>      <th>  AIC:               </th> <td>  -296.5</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>  1014</td>      <th>  BIC:               </th> <td>  -227.4</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>    13</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>       <td>cluster</td>     <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "             <td></td>               <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th>              <td>-8.148e-09</td> <td> 1.44e-08</td> <td>   -0.565</td> <td> 0.572</td> <td>-3.64e-08</td> <td> 2.01e-08</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_age</th>             <td>  4.48e-17</td> <td> 2.65e-18</td> <td>   16.927</td> <td> 0.000</td> <td> 3.96e-17</td> <td>    5e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_asq</th>             <td> 1.008e-14</td> <td> 6.82e-16</td> <td>   14.766</td> <td> 0.000</td> <td> 8.74e-15</td> <td> 1.14e-14</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_bmi</th>             <td> 1.256e-17</td> <td>  7.3e-18</td> <td>    1.722</td> <td> 0.085</td> <td>-1.74e-18</td> <td> 2.69e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_hispanic</th>        <td> 1.353e-17</td> <td> 5.98e-18</td> <td>    2.265</td> <td> 0.024</td> <td> 1.82e-18</td> <td> 2.52e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_black</th>           <td> 6.053e-18</td> <td> 6.48e-18</td> <td>    0.935</td> <td> 0.350</td> <td>-6.64e-18</td> <td> 1.87e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_other</th>           <td> -3.41e-17</td> <td> 1.86e-17</td> <td>   -1.832</td> <td> 0.067</td> <td>-7.06e-17</td> <td> 2.37e-18</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_asian</th>           <td> 2.047e-18</td> <td> 9.81e-18</td> <td>    0.209</td> <td> 0.835</td> <td>-1.72e-17</td> <td> 2.13e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_schooling</th>       <td> 1.101e-17</td> <td> 1.06e-17</td> <td>    1.042</td> <td> 0.298</td> <td>-9.71e-18</td> <td> 3.17e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_cohab</th>           <td>-2.308e-17</td> <td> 1.16e-17</td> <td>   -1.982</td> <td> 0.047</td> <td>-4.59e-17</td> <td>-2.58e-19</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_married</th>         <td> 4.384e-17</td> <td> 1.25e-17</td> <td>    3.511</td> <td> 0.000</td> <td> 1.94e-17</td> <td> 6.83e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_divorced</th>        <td> 2.291e-17</td> <td> 7.96e-18</td> <td>    2.878</td> <td> 0.004</td> <td> 7.31e-18</td> <td> 3.85e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_separated</th>       <td> 9.066e-19</td> <td> 9.29e-18</td> <td>    0.098</td> <td> 0.922</td> <td>-1.73e-17</td> <td> 1.91e-17</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_age_cl</th>          <td>    0.0023</td> <td>    0.007</td> <td>    0.338</td> <td> 0.735</td> <td>   -0.011</td> <td>    0.016</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_unsafe</th>          <td>    0.0510</td> <td>    0.028</td> <td>    1.794</td> <td> 0.073</td> <td>   -0.005</td> <td>    0.107</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_llength</th>         <td>   -0.4345</td> <td>    0.024</td> <td>  -17.758</td> <td> 0.000</td> <td>   -0.482</td> <td>   -0.387</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_reg</th>             <td>   -0.0373</td> <td>    0.019</td> <td>   -1.979</td> <td> 0.048</td> <td>   -0.074</td> <td>   -0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_asq_cl</th>          <td> -1.47e-05</td> <td>  7.6e-05</td> <td>   -0.194</td> <td> 0.847</td> <td>   -0.000</td> <td>    0.000</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_appearance_cl</th>   <td>    0.0056</td> <td>    0.006</td> <td>    0.965</td> <td> 0.334</td> <td>   -0.006</td> <td>    0.017</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_provider_second</th> <td>    0.1131</td> <td>    0.061</td> <td>    1.862</td> <td> 0.063</td> <td>   -0.006</td> <td>    0.232</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_asian_cl</th>        <td>   -0.0099</td> <td>    0.034</td> <td>   -0.290</td> <td> 0.772</td> <td>   -0.077</td> <td>    0.057</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_black_cl</th>        <td>    0.0265</td> <td>    0.043</td> <td>    0.622</td> <td> 0.534</td> <td>   -0.057</td> <td>    0.110</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_hispanic_cl</th>     <td>   -0.0621</td> <td>    0.052</td> <td>   -1.193</td> <td> 0.233</td> <td>   -0.164</td> <td>    0.040</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_othrace_cl</th>      <td>    0.1422</td> <td>    0.050</td> <td>    2.867</td> <td> 0.004</td> <td>    0.045</td> <td>    0.239</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_hot</th>             <td>    0.0525</td> <td>    0.027</td> <td>    1.924</td> <td> 0.054</td> <td>   -0.001</td> <td>    0.106</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>demean_massage_cl</th>      <td>   -0.0010</td> <td>    0.028</td> <td>   -0.037</td> <td> 0.970</td> <td>   -0.056</td> <td>    0.054</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td>79.779</td> <th>  Durbin-Watson:     </th> <td>   2.528</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 361.895</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.172</td> <th>  Prob(JB):          </th> <td>2.60e-79</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 5.886</td> <th>  Cond. No.          </th> <td>1.00e+16</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors are robust to cluster correlation (cluster)<br/>[2] The smallest eigenvalue is 6.38e-24. This might indicate that there are<br/>strong multicollinearity problems or that the design matrix is singular."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:             demean_lnw   R-squared:                       0.516\n",
+       "Model:                            OLS   Adj. R-squared:                  0.510\n",
+       "Method:                 Least Squares   F-statistic:                     32.39\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           9.91e-47\n",
+       "Time:                        13:32:50   Log-Likelihood:                 162.25\n",
+       "No. Observations:                1028   AIC:                            -296.5\n",
+       "Df Residuals:                    1014   BIC:                            -227.4\n",
+       "Df Model:                          13                                         \n",
+       "Covariance Type:              cluster                                         \n",
+       "==========================================================================================\n",
+       "                             coef    std err          z      P>|z|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------------------\n",
+       "Intercept              -8.148e-09   1.44e-08     -0.565      0.572   -3.64e-08    2.01e-08\n",
+       "demean_age               4.48e-17   2.65e-18     16.927      0.000    3.96e-17       5e-17\n",
+       "demean_asq              1.008e-14   6.82e-16     14.766      0.000    8.74e-15    1.14e-14\n",
+       "demean_bmi              1.256e-17    7.3e-18      1.722      0.085   -1.74e-18    2.69e-17\n",
+       "demean_hispanic         1.353e-17   5.98e-18      2.265      0.024    1.82e-18    2.52e-17\n",
+       "demean_black            6.053e-18   6.48e-18      0.935      0.350   -6.64e-18    1.87e-17\n",
+       "demean_other            -3.41e-17   1.86e-17     -1.832      0.067   -7.06e-17    2.37e-18\n",
+       "demean_asian            2.047e-18   9.81e-18      0.209      0.835   -1.72e-17    2.13e-17\n",
+       "demean_schooling        1.101e-17   1.06e-17      1.042      0.298   -9.71e-18    3.17e-17\n",
+       "demean_cohab           -2.308e-17   1.16e-17     -1.982      0.047   -4.59e-17   -2.58e-19\n",
+       "demean_married          4.384e-17   1.25e-17      3.511      0.000    1.94e-17    6.83e-17\n",
+       "demean_divorced         2.291e-17   7.96e-18      2.878      0.004    7.31e-18    3.85e-17\n",
+       "demean_separated        9.066e-19   9.29e-18      0.098      0.922   -1.73e-17    1.91e-17\n",
+       "demean_age_cl              0.0023      0.007      0.338      0.735      -0.011       0.016\n",
+       "demean_unsafe              0.0510      0.028      1.794      0.073      -0.005       0.107\n",
+       "demean_llength            -0.4345      0.024    -17.758      0.000      -0.482      -0.387\n",
+       "demean_reg                -0.0373      0.019     -1.979      0.048      -0.074      -0.000\n",
+       "demean_asq_cl           -1.47e-05    7.6e-05     -0.194      0.847      -0.000       0.000\n",
+       "demean_appearance_cl       0.0056      0.006      0.965      0.334      -0.006       0.017\n",
+       "demean_provider_second     0.1131      0.061      1.862      0.063      -0.006       0.232\n",
+       "demean_asian_cl           -0.0099      0.034     -0.290      0.772      -0.077       0.057\n",
+       "demean_black_cl            0.0265      0.043      0.622      0.534      -0.057       0.110\n",
+       "demean_hispanic_cl        -0.0621      0.052     -1.193      0.233      -0.164       0.040\n",
+       "demean_othrace_cl          0.1422      0.050      2.867      0.004       0.045       0.239\n",
+       "demean_hot                 0.0525      0.027      1.924      0.054      -0.001       0.106\n",
+       "demean_massage_cl         -0.0010      0.028     -0.037      0.970      -0.056       0.054\n",
+       "==============================================================================\n",
+       "Omnibus:                       79.779   Durbin-Watson:                   2.528\n",
+       "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              361.895\n",
+       "Skew:                           0.172   Prob(JB):                     2.60e-79\n",
+       "Kurtosis:                       5.886   Cond. No.                     1.00e+16\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors are robust to cluster correlation (cluster)\n",
+       "[2] The smallest eigenvalue is 6.38e-24. This might indicate that there are\n",
+       "strong multicollinearity problems or that the design matrix is singular.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 12,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#-- Demean OLS\n",
+    "dm_formula = \"\"\"demean_lnw ~ demean_age + demean_asq + demean_bmi + \n",
+    "                demean_hispanic + demean_black + demean_other +\n",
+    "                demean_asian + demean_schooling + demean_cohab + \n",
+    "                demean_married + demean_divorced + demean_separated +\n",
+    "                demean_age_cl + demean_unsafe + demean_llength + demean_reg + \n",
+    "                demean_asq_cl + demean_appearance_cl + \n",
+    "                demean_provider_second + demean_asian_cl + demean_black_cl + \n",
+    "                demean_hispanic_cl + demean_othrace_cl +\n",
+    "                demean_hot + demean_massage_cl\"\"\"\n",
+    "\n",
+    "ols = sm.OLS.from_formula(dm_formula, data=balanced_sasp).fit(cov_type='cluster', \n",
+    "                                                           cov_kwds={'groups': balanced_sasp['id']})\n",
+    "ols.summary()  \n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### QUESTIONS\n",
+    "- Interpret the effect of natural log of session length on the natural log of hourly wage.  Describe the economic theory that might explain this relationship?  (HINT: Consider the role that supplier fixed versus variable costs may have on the hourly wage.)\n",
+    "- Becker described discrimination in terms of ``taste based``.  This meant that social interactions with people of the other race were factors into marginal cost.  Given that these persist, what does this imply about the effect that competition is having on discrimination?\n",
+    "- Hamermesh and Biddle suggest that beauty is valued on the market.  Describe some reasons why there is no effect on client beauty once we use the within estimators?\n",
+    "- What other interesting results did you find in this analysis?  Which ones surprised you and which ones were intuitive and why?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Python/Potential_Outcomes.ipynb b/Python/Potential_Outcomes.ipynb
new file mode 100644
index 0000000..be962a8
--- /dev/null
+++ b/Python/Potential_Outcomes.ipynb
@@ -0,0 +1,493 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Potential Outcomes** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "from itertools import combinations\n",
+    "import plotnine as p"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_stata(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>          <td>paup</td>       <th>  R-squared:         </th> <td>   0.697</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.665</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   21.49</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th> <td>2.00e-07</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:33:10</td>     <th>  Log-Likelihood:    </th> <td> -115.47</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>    32</td>      <th>  AIC:               </th> <td>   238.9</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>    28</td>      <th>  BIC:               </th> <td>   244.8</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     3</td>      <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th> <td>   63.1877</td> <td>   27.144</td> <td>    2.328</td> <td> 0.027</td> <td>    7.586</td> <td>  118.789</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>outrelief</th> <td>    0.7521</td> <td>    0.135</td> <td>    5.572</td> <td> 0.000</td> <td>    0.476</td> <td>    1.029</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>old</th>       <td>    0.0556</td> <td>    0.223</td> <td>    0.249</td> <td> 0.805</td> <td>   -0.402</td> <td>    0.513</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>pop</th>       <td>   -0.3107</td> <td>    0.067</td> <td>   -4.648</td> <td> 0.000</td> <td>   -0.448</td> <td>   -0.174</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td> 7.594</td> <th>  Durbin-Watson:     </th> <td>   2.344</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.022</td> <th>  Jarque-Bera (JB):  </th> <td>   5.979</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.961</td> <th>  Prob(JB):          </th> <td>  0.0503</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 3.888</td> <th>  Cond. No.          </th> <td>2.56e+03</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 2.56e+03. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                   paup   R-squared:                       0.697\n",
+       "Model:                            OLS   Adj. R-squared:                  0.665\n",
+       "Method:                 Least Squares   F-statistic:                     21.49\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):           2.00e-07\n",
+       "Time:                        13:33:10   Log-Likelihood:                -115.47\n",
+       "No. Observations:                  32   AIC:                             238.9\n",
+       "Df Residuals:                      28   BIC:                             244.8\n",
+       "Df Model:                           3                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "==============================================================================\n",
+       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept     63.1877     27.144      2.328      0.027       7.586     118.789\n",
+       "outrelief      0.7521      0.135      5.572      0.000       0.476       1.029\n",
+       "old            0.0556      0.223      0.249      0.805      -0.402       0.513\n",
+       "pop           -0.3107      0.067     -4.648      0.000      -0.448      -0.174\n",
+       "==============================================================================\n",
+       "Omnibus:                        7.594   Durbin-Watson:                   2.344\n",
+       "Prob(Omnibus):                  0.022   Jarque-Bera (JB):                5.979\n",
+       "Skew:                           0.961   Prob(JB):                       0.0503\n",
+       "Kurtosis:                       3.888   Cond. No.                     2.56e+03\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "[2] The condition number is large, 2.56e+03. This might indicate that there are\n",
+       "strong multicollinearity or other numerical problems.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "yule = read_data('yule.dta')\n",
+    "\n",
+    "res = sm.OLS.from_formula('paup ~ outrelief + old + pop', yule).fit()\n",
+    "res.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "- How do you interpret the coefficient on `outrelief` given it's a percentage change regressed onto a percentage?\n",
+    "- Draw a DAG representing what must be true in order for Yule's estimate of `outrelief` on pauper growth rates to be causal?  \n",
+    "- Yule concluded that public assistance (`outrelief`) increased pauper growth rates. How convinced are you that all backdoor paths between pauperism and out-relief are blocked once you control for two covariates in a cross-sectional database for all of England? Could there be unobserved determinants of both poverty and public assistance?\n",
+    "- If public assistance causes pauper growth rates, but pauper growth rates also causes public assistance, then why won't Yule's regression capture a causal effect of `outrelief` on pauper growth rates?  Explain the concept of reverse causality with Yule's data."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## Independence Assumption"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.617"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def gap():\n",
+    "    sdo = pd.DataFrame({\n",
+    "        'y1': (7, 5, 5, 7, 4, 10, 1, 5, 3, 9),\n",
+    "        'y0' : (1, 6, 1, 8, 2, 1, 10, 6, 7, 8),\n",
+    "        'random' : np.random.normal(size=10)})\n",
+    "    sdo.sort_values('random', inplace=True)\n",
+    "    sdo['d'] = [1,1,1,1,1,0,0,0,0,0]\n",
+    "    sdo['y'] = sdo['d']*sdo['y1'] + (1-sdo['d'])*sdo['y0']\n",
+    "\n",
+    "    sdo = np.mean(sdo.y.values[0:5] - sdo.y.values[5:10])\n",
+    "\n",
+    "    return sdo\n",
+    "\n",
+    "\n",
+    "sim = [gap() for x in range(1000)]\n",
+    "np.mean(sim)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- The requirement that treatment be independent of potential outcomes states that a choice made by a person must be independent of what they expect to gain or lose from the choice.  Give an example where this is likely true?  What does independence imply about human decision-making?\n",
+    "- All of the behavioral sciences, including economics, suggest that independence is unlikely to hold outside of an experiment. What is so special about an experiment where independence will hold?  What is so special about behavior outside an experiment where it is unlikely to hold?\n",
+    "- What implication does the decision rule of utility maximization from economics have for our ability to appeal to treatment being distributed independent of potential outcomes?\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "\n",
+    "## Fisher Randomization"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.014285714285714285"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "correct = pd.DataFrame({'cup': np.arange(1,9), \n",
+    "                        'guess':np.concatenate((range(1,5), np.repeat(0, 4)))})\n",
+    "\n",
+    "combo = pd.DataFrame(np.array(list(combinations(correct['cup'], 4))), \n",
+    "                     columns=['cup_1', 'cup_2', 'cup_3', 'cup_4'])\n",
+    "combo['permutation'] = np.arange(70)\n",
+    "combo['key'] = 1\n",
+    "correct['key'] = 1\n",
+    "combo = pd.merge(correct, combo, on='key')\n",
+    "combo.drop('key', axis=1, inplace=True)\n",
+    "combo['correct'] = 0\n",
+    "combo.loc[(combo.cup_1==1) & \n",
+    "          (combo.cup_2==2) & \n",
+    "          (combo.cup_3==3) & \n",
+    "          (combo.cup_4==4), 'correct'] = 1\n",
+    "combo = combo.sort_values(['permutation', 'cup'])\n",
+    "\n",
+    "p_value = combo.correct.sum()/combo.shape[0]\n",
+    "p_value"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "- Using the above simulation, what is the probability that Dr. Bristol selected the correct four cups completely by chance?\n",
+    "\n",
+    "## Randomization Inference\n",
+    "\n",
+    "### Fisher Sharp Null"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0    0.585714\n",
+       "Name: rank, dtype: float64"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "ri = read_data('ri.dta')\n",
+    "ri['id'] = range(1,9)\n",
+    "treated = range(1,5)\n",
+    "\n",
+    "combo = pd.DataFrame(np.array(list(combinations(ri['id'], 4))), \n",
+    "                     columns=['treated1', 'treated2', 'treated3', 'treated4'])\n",
+    "combo['permutation'] = np.arange(1,71)\n",
+    "\n",
+    "combo['key'] = 1\n",
+    "ri['key'] = 1\n",
+    "combo = pd.merge(ri, combo, on='key')\n",
+    "combo.drop('key', axis=1, inplace=True)\n",
+    "combo = combo.sort_values(['permutation', 'name'])\n",
+    "\n",
+    "combo['d'] = 0\n",
+    "combo.loc[(combo.treated1==combo.id) | \n",
+    "          (combo.treated2==combo.id) | \n",
+    "          (combo.treated3==combo.id) | \n",
+    "          (combo.treated4==combo.id), 'd'] = 1\n",
+    "\n",
+    "te1 = combo[combo.d==1].groupby('permutation')['y'].mean()\n",
+    "te0 = combo[combo.d==0].groupby('permutation')['y'].mean()\n",
+    "\n",
+    "n = pd.merge(te1, te0, how='inner', on=\"permutation\").shape[0]\n",
+    "\n",
+    "p_value = pd.merge(te1, te0, how='inner', on=\"permutation\")\n",
+    "p_value.columns = ['te1', 'te0']\n",
+    "p_value = p_value.reset_index()\n",
+    "p_value['ate'] = p_value['te1'] - p_value['te0']\n",
+    "p_value = p_value.sort_values(['ate', 'permutation'])\n",
+    "p_value['rank'] = range(1, p_value.shape[0]+1)\n",
+    "p_value = p_value[p_value['permutation'] == 1]\n",
+    "p_value['rank'] / n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "#### Questions\n",
+    "\n",
+    "- Can we reject the null in the placebo distribution?\n",
+    "\n",
+    "\n",
+    "\n",
+    "### KS Test\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8764080721288)>"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tb = pd.DataFrame({\n",
+    "    'd' : np.concatenate((np.repeat(0, 20), np.repeat(1, 20))),\n",
+    "    'y' : (\n",
+    "        0.22, -0.87, -2.39, -1.79, 0.37, -1.54,\n",
+    "        1.28, -0.31, -0.74, 1.72,\n",
+    "        0.38, -0.17, -0.62, -1.10, 0.30,\n",
+    "        0.15, 2.30, 0.19, -0.50, -0.9,\n",
+    "        -5.13, -2.19, 2.43, -3.83, 0.5,\n",
+    "        -3.25, 4.32, 1.63, 5.18, -0.43,\n",
+    "        7.11, 4.87, -3.10, -5.81, 3.76,\n",
+    "        6.31, 2.58, 0.07, 5.76, 3.50\n",
+    "    )})\n",
+    "\n",
+    "p.ggplot() +\\\n",
+    "    p.geom_density(tb, p.aes(x='y', color='factor(d)')) +\\\n",
+    "    p.xlim(-7, 8) +\\\n",
+    "    p.labs(title = \"Kolmogorov-Smirnov Test\") +\\\n",
+    "    p.scale_color_discrete(labels = (\"Control\", \"Treatment\"))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Approximate $p$-values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.4505518674850464\n",
+      "0    0.001\n",
+      "Name: rank, dtype: float64\n"
+     ]
+    }
+   ],
+   "source": [
+    "hiv = read_data(\"thornton_hiv.dta\")\n",
+    "# creating the permutations\n",
+    "\n",
+    "def permuteHIV(df, random = True):\n",
+    "    tb = df.copy()\n",
+    "    first_half = np.ceil(tb.shape[0] / 2)\n",
+    "    second_half = tb.shape[0] - first_half\n",
+    "    if random:\n",
+    "        tb = tb.sample(frac=1)\n",
+    "        tb['any'] = np.concatenate((np.repeat(1, first_half), np.repeat(0, second_half)))\n",
+    "    \n",
+    "    te1 = tb[tb['any']==1]['got'].mean()\n",
+    "    te0 = tb[tb['any']==0]['got'].mean()\n",
+    "    \n",
+    "    \n",
+    "    ate = te1 - te0\n",
+    "    return ate\n",
+    "\n",
+    "print(permuteHIV(hiv, random = False))\n",
+    "iterations = 1000\n",
+    "permutation = pd.DataFrame({\n",
+    "    'iteration': range(iterations),\n",
+    "    'ate' : [permuteHIV(hiv, random=False), \n",
+    "                            *[permuteHIV(hiv, random=True) for x in range(iterations-1)]]}\n",
+    ")\n",
+    "# calculating the p-value\n",
+    "\n",
+    "permutation = permutation.sort_values('ate', ascending=False)\n",
+    "permutation['rank'] = np.arange(1, iterations+1)\n",
+    "p_value = permutation[permutation.iteration==0]['rank'].astype(float) / iterations\n",
+    "print(p_value)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions \n",
+    "\n",
+    "- How does the randomization inference test of no treatment effect differ from a null of no average treatment effect?\n",
+    "- How likely is it that Thornton's results were a result of random chance? "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Python/Probability_and_Regression.ipynb b/Python/Probability_and_Regression.ipynb
new file mode 100644
index 0000000..d966418
--- /dev/null
+++ b/Python/Probability_and_Regression.ipynb
@@ -0,0 +1,1116 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Probability and Regression** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n",
+    "\n",
+    "### Packages needed\n",
+    "\n",
+    "The first thing you need to do is install a few packages to make sure everything runs:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "\n",
+    "import plotnine as p"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_csv(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## OLS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "np.random.seed(1)\n",
+    "\n",
+    "tb = pd.DataFrame({\n",
+    "    'x': np.random.normal(size=10000),\n",
+    "    'u': np.random.normal(size=10000)})\n",
+    "tb['y'] = 5.5*tb['x'].values + 12*tb['u'].values"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>            <td>y</td>        <th>  R-squared:         </th> <td>   0.183</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.183</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   2237.</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:33:44</td>     <th>  Log-Likelihood:    </th> <td> -39049.</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td> 10000</td>      <th>  AIC:               </th> <td>7.810e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>  9998</td>      <th>  BIC:               </th> <td>7.812e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th> <td>    0.1114</td> <td>    0.120</td> <td>    0.927</td> <td> 0.354</td> <td>   -0.124</td> <td>    0.347</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>x</th>         <td>    5.6887</td> <td>    0.120</td> <td>   47.293</td> <td> 0.000</td> <td>    5.453</td> <td>    5.924</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td> 0.640</td> <th>  Durbin-Watson:     </th> <td>   2.050</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.726</td> <th>  Jarque-Bera (JB):  </th> <td>   0.672</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td>-0.012</td> <th>  Prob(JB):          </th> <td>   0.715</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 2.968</td> <th>  Cond. No.          </th> <td>    1.01</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                      y   R-squared:                       0.183\n",
+       "Model:                            OLS   Adj. R-squared:                  0.183\n",
+       "Method:                 Least Squares   F-statistic:                     2237.\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):               0.00\n",
+       "Time:                        13:33:44   Log-Likelihood:                -39049.\n",
+       "No. Observations:               10000   AIC:                         7.810e+04\n",
+       "Df Residuals:                    9998   BIC:                         7.812e+04\n",
+       "Df Model:                           1                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "==============================================================================\n",
+       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept      0.1114      0.120      0.927      0.354      -0.124       0.347\n",
+       "x              5.6887      0.120     47.293      0.000       5.453       5.924\n",
+       "==============================================================================\n",
+       "Omnibus:                        0.640   Durbin-Watson:                   2.050\n",
+       "Prob(Omnibus):                  0.726   Jarque-Bera (JB):                0.672\n",
+       "Skew:                          -0.012   Prob(JB):                        0.715\n",
+       "Kurtosis:                       2.968   Cond. No.                         1.01\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "reg_tb = sm.OLS.from_formula('y ~ x', data=tb).fit()\n",
+    "reg_tb.summary()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "tb['yhat1'] = reg_tb.predict(tb)\n",
+    "tb['yhat2'] = 0.1114 + 5.6887*tb['x']\n",
+    "tb['uhat1'] = reg_tb.resid\n",
+    "tb['uhat2'] = tb['y'] - tb['yhat2']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>x</th>\n",
+       "      <th>u</th>\n",
+       "      <th>y</th>\n",
+       "      <th>yhat1</th>\n",
+       "      <th>yhat2</th>\n",
+       "      <th>uhat1</th>\n",
+       "      <th>uhat2</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>1.624345</td>\n",
+       "      <td>-0.122474</td>\n",
+       "      <td>7.464213</td>\n",
+       "      <td>9.351735</td>\n",
+       "      <td>9.351813</td>\n",
+       "      <td>-1.887522</td>\n",
+       "      <td>-1.887601</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>-0.611756</td>\n",
+       "      <td>0.228170</td>\n",
+       "      <td>-0.626622</td>\n",
+       "      <td>-3.368695</td>\n",
+       "      <td>-3.368699</td>\n",
+       "      <td>2.742073</td>\n",
+       "      <td>2.742076</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>-0.528172</td>\n",
+       "      <td>-0.352305</td>\n",
+       "      <td>-7.132606</td>\n",
+       "      <td>-2.893210</td>\n",
+       "      <td>-2.893211</td>\n",
+       "      <td>-4.239396</td>\n",
+       "      <td>-4.239396</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>-1.072969</td>\n",
+       "      <td>-0.830553</td>\n",
+       "      <td>-15.867969</td>\n",
+       "      <td>-5.992376</td>\n",
+       "      <td>-5.992397</td>\n",
+       "      <td>-9.875593</td>\n",
+       "      <td>-9.875572</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>0.865408</td>\n",
+       "      <td>-0.261090</td>\n",
+       "      <td>1.626664</td>\n",
+       "      <td>5.034394</td>\n",
+       "      <td>5.034444</td>\n",
+       "      <td>-3.407730</td>\n",
+       "      <td>-3.407780</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "          x         u          y     yhat1     yhat2     uhat1     uhat2\n",
+       "0  1.624345 -0.122474   7.464213  9.351735  9.351813 -1.887522 -1.887601\n",
+       "1 -0.611756  0.228170  -0.626622 -3.368695 -3.368699  2.742073  2.742076\n",
+       "2 -0.528172 -0.352305  -7.132606 -2.893210 -2.893211 -4.239396 -4.239396\n",
+       "3 -1.072969 -0.830553 -15.867969 -5.992376 -5.992397 -9.875593 -9.875572\n",
+       "4  0.865408 -0.261090   1.626664  5.034394  5.034444 -3.407730 -3.407780"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tb.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122428651)>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(tb, p.aes(x='x', y='y')) +\\\n",
+    "    p.ggtitle(\"OLS Regression Line\") +\\\n",
+    "    p.geom_point(size = 0.05, color = \"black\", alpha = 0.5) +\\\n",
+    "    p.geom_smooth(p.aes(x='x', y='y'), method = \"lm\", color = \"black\") +\\\n",
+    "    p.annotate(\"text\", x = -1.5, y = 30, color = \"red\", \n",
+    "             label = \"Intercept = {}\".format(-0.0732608)) +\\\n",
+    "    p.annotate(\"text\", x = 1.5, y = -30, color = \"blue\", \n",
+    "             label = \"Slope = {}\".format(5.685033))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions:\n",
+    "\n",
+    "-   What is the predicted value of $y$ when $x = 0$?\n",
+    "-   How much do we estimate $y$ increases by when $x$ increases by one unit?\n",
+    "-   Assume we y was the natural log of some variable, and x was the natural log of some variable.  How do we interpret the coefficient on $x$ if it is a log-log regression?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>x</th>\n",
+       "      <th>u</th>\n",
+       "      <th>y</th>\n",
+       "      <th>yhat1</th>\n",
+       "      <th>uhat1</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>-17.384075</td>\n",
+       "      <td>-39.860946</td>\n",
+       "      <td>-131.874117</td>\n",
+       "      <td>-90.334179</td>\n",
+       "      <td>-41.539938</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>-1.443835</td>\n",
+       "      <td>31.905182</td>\n",
+       "      <td>59.478859</td>\n",
+       "      <td>9.498051</td>\n",
+       "      <td>49.980808</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>-7.489976</td>\n",
+       "      <td>-1.280647</td>\n",
+       "      <td>-25.031221</td>\n",
+       "      <td>-28.368362</td>\n",
+       "      <td>3.337141</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>-1.855004</td>\n",
+       "      <td>11.358450</td>\n",
+       "      <td>17.151889</td>\n",
+       "      <td>6.922939</td>\n",
+       "      <td>10.228950</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>9.236294</td>\n",
+       "      <td>-31.584772</td>\n",
+       "      <td>-35.460660</td>\n",
+       "      <td>76.386702</td>\n",
+       "      <td>-111.847362</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           x          u           y      yhat1       uhat1\n",
+       "0 -17.384075 -39.860946 -131.874117 -90.334179  -41.539938\n",
+       "1  -1.443835  31.905182   59.478859   9.498051   49.980808\n",
+       "2  -7.489976  -1.280647  -25.031221 -28.368362    3.337141\n",
+       "3  -1.855004  11.358450   17.151889   6.922939   10.228950\n",
+       "4   9.236294 -31.584772  -35.460660  76.386702 -111.847362"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tb = pd.DataFrame({\n",
+    "    'x': 9*np.random.normal(size=10),\n",
+    "    'u': 36*np.random.normal(size=10)})\n",
+    "tb['y'] = 3*tb['x'].values + 2*tb['u'].values\n",
+    "\n",
+    "reg_tb = sm.OLS.from_formula('y ~ x', data=tb).fit()\n",
+    "\n",
+    "tb['yhat1'] = reg_tb.predict(tb)\n",
+    "tb['uhat1'] = reg_tb.resid\n",
+    "\n",
+    "tb.head()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "-   What is the average of the residuals $\\hat{u}$ from our regression?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "coefs = np.zeros(1000)\n",
+    "for i in range(1000):\n",
+    "    tb = pd.DataFrame({\n",
+    "    'x': 9*np.random.normal(size=10000),\n",
+    "    'u': 36*np.random.normal(size=10000)})\n",
+    "    tb['y'] = 3 + 2*tb['x'].values + tb['u'].values\n",
+    "\n",
+    "    reg_tb = sm.OLS.from_formula('y ~ x', data=tb).fit()\n",
+    "\n",
+    "    coefs[i] = reg_tb.params['x']"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122428624)>"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot() +\\\n",
+    "  p.geom_histogram(p.aes(x=coefs), binwidth = 0.01)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "-   Explain the concept of unbiasedness in the context of this simulation?  \n",
+    "-   On average, do we think the estimate is close to the true value of $\\beta_1 = 2$?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/smoothers.py:310: PlotnineWarning: Confidence intervals are not yet implementedfor lowess smoothings.\n",
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/smoothers.py:310: PlotnineWarning: Confidence intervals are not yet implementedfor lowess smoothings.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122631054)>"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "auto = pd.read_stata('http://www.stata-press.com/data/r13/auto.dta')\n",
+    "auto['length'] = auto['length'] - auto['length'].mean()\n",
+    "\n",
+    "lm1 = sm.OLS.from_formula('price ~ length', data=auto).fit()\n",
+    "lm2 = sm.OLS.from_formula('price ~ length + weight + headroom + mpg', data=auto).fit()\n",
+    "\n",
+    "\n",
+    "coef_lm1 = lm1.params\n",
+    "coef_lm2 = lm2.params\n",
+    "resid_lm2 = lm2.resid\n",
+    "\n",
+    "auto['y_single'] = coef_lm1[0] + coef_lm1[1]*auto['length']\n",
+    "auto['y_multi'] = coef_lm1[0] + coef_lm2[1]*auto['length']\n",
+    "\n",
+    "p.ggplot(auto) +\\\n",
+    "  p.geom_point(p.aes(x = 'length', y = 'price')) +\\\n",
+    "  p.geom_smooth(p.aes(x = 'length', y = 'y_multi'), color = \"blue\") +\\\n",
+    "  p.geom_smooth(p.aes(x = 'length', y = 'y_single'), color=\"red\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "\n",
+    "-   What happened to the coefficient on length after controlling for weight, headroom, and mpg in the regression?\n",
+    "\n",
+    "## Clustering Standard Errors\n",
+    "\n",
+    "### Cluster robust standard errors\n",
+    "\n",
+    "People will try to scare you by challenging how you constructed your standard errors. Heteroskedastic errors, though, aren't the only thing you should be worried about when it comes to inference. Some phenomena do not affect observations individually, but they do affect groups of observations that involve individuals. And then they affect those individuals within the group in a common way. Say you want to estimate the effect of class size on student achievement, but you know that there exist unobservable things (like the teacher) that affect all the students equally. If we can commit to independence of these unobservables across classes, but individual student unobservables are correlated within a class, then we have a situation in which we need to cluster the standard errors. Before we dive into an example, I'd like to start with a simulation to illustrate the problem.\n",
+    "\n",
+    "As a baseline for this simulation, let's begin by simulating nonclustered data and analyze least squares estimates of that nonclustered data. This will help firm up our understanding of the problems that occur with least squares when data is clustered.\n",
+    "\n",
+    "First, I will create a function to generate our Monte Carlo simulation."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def gen_cluster(param = (.1, .5), n = 1000, n_cluster = 50, rho = .5):\n",
+    "    # Function to generate clustered data\n",
+    "\n",
+    "    # individual level\n",
+    "    Sigma_i = np.array((1, 0, 0, 1 - rho)).reshape(2,2)\n",
+    "\n",
+    "    values_i = np.random.multivariate_normal(np.zeros(2), Sigma_i, size = n)\n",
+    "\n",
+    "    # cluster level\n",
+    "    cluster_name = np.repeat(np.arange(1, n_cluster+1), repeats = n / n_cluster)\n",
+    "    Sigma_cl = np.array((1, 0, 0, rho)).reshape(2,2)\n",
+    "    values_cl = np.random.multivariate_normal(np.zeros(2),Sigma_cl, size = n_cluster)\n",
+    "\n",
+    "    # predictor var consists of individual- and cluster-level components\n",
+    "    x = values_i[: , 0] + np.repeat(values_cl[: , 0], repeats = n / n_cluster)\n",
+    "\n",
+    "    # error consists of individual- and cluster-level components\n",
+    "    error = values_i[: , 1] + np.repeat(values_cl[: , 1], repeats = n / n_cluster)\n",
+    "\n",
+    "    # data generating process\n",
+    "    y = param[0] + param[1]*x + error\n",
+    "    \n",
+    "    df = pd.DataFrame({'x':x, 'y':y, 'cluster': cluster_name})\n",
+    "    return df\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def cluster_sim(param = (.1, .5), n = 1000, n_cluster = 50,\n",
+    "                        rho = .5, cluster_robust = False):\n",
+    "\n",
+    "    df = gen_cluster(param = param, n = n , n_cluster = n_cluster, rho = rho)\n",
+    "\n",
+    "    if not cluster_robust:\n",
+    "        fit = sm.OLS.from_formula('y ~ x', data = df).fit()\n",
+    "    else: # cluster-robust SE\n",
+    "        fit = sm.OLS.from_formula('y ~ x', data = df).fit(cov_type='cluster', cov_kwds={'groups': df['cluster']})\n",
+    "    \n",
+    "    b1 = fit.params[1]\n",
+    "    Sigma = fit.cov_params()   \n",
+    "    \n",
+    "    se = np.sqrt(np.diag(Sigma)[1])\n",
+    "    ci95 = se*1.96\n",
+    "    b1_ci95 = (b1-ci95, b1+ci95)\n",
+    "\n",
+    "\n",
+    "    return (b1, se, *b1_ci95)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n_sims = 1000\n",
+    "param = (.1, .5)\n",
+    "n = 1000\n",
+    "n_cluster = 50\n",
+    "rho = .5\n",
+    "cluster_robust = True"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def run_cluster_sim(n_sims = 1000, param = (.1, .5), n = 1000,\n",
+    "                            n_cluster = 50, rho = .5, cluster_robust = False):\n",
+    "\n",
+    "    res = [cluster_sim(param = param, n = n, rho = rho,\n",
+    "                                      n_cluster = n_cluster,\n",
+    "                                      cluster_robust = cluster_robust) for x in range(n_sims)]\n",
+    "    df = pd.DataFrame(res)\n",
+    "    df.columns = ('b1', 'se_b1', 'ci95_lower', 'ci95_upper')\n",
+    "    df['param_caught'] = (df['ci95_lower'] <= param[1]) & (param[1] <= df['ci95_upper'])\n",
+    "    df['id'] = df.index\n",
+    "    return df\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Simulation no clustered SE\n",
+    "sim_params = [.4, 0] # beta1 = 0: no effect of x on y\n",
+    "sim_nocluster = run_cluster_sim(n_sims=1000, param = sim_params, rho=0, cluster_robust = False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/stat_bin.py:93: PlotnineWarning: 'stat_bin()' using 'bins = 28'. Pick better value with 'binwidth'.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122630455)>"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(sim_nocluster, p.aes('b1')) +\\\n",
+    "  p.geom_histogram(color = 'black') +\\\n",
+    "  p.geom_vline(xintercept = sim_params[1], color = 'red')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>b1</th>\n",
+       "      <th>se_b1</th>\n",
+       "      <th>ci95_lower</th>\n",
+       "      <th>ci95_upper</th>\n",
+       "      <th>param_caught</th>\n",
+       "      <th>id</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>765</th>\n",
+       "      <td>-0.079756</td>\n",
+       "      <td>0.023947</td>\n",
+       "      <td>-0.126692</td>\n",
+       "      <td>-0.032821</td>\n",
+       "      <td>False</td>\n",
+       "      <td>765</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>29</th>\n",
+       "      <td>-0.054530</td>\n",
+       "      <td>0.021139</td>\n",
+       "      <td>-0.095962</td>\n",
+       "      <td>-0.013098</td>\n",
+       "      <td>False</td>\n",
+       "      <td>29</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>782</th>\n",
+       "      <td>-0.039275</td>\n",
+       "      <td>0.022142</td>\n",
+       "      <td>-0.082672</td>\n",
+       "      <td>0.004123</td>\n",
+       "      <td>True</td>\n",
+       "      <td>782</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>998</th>\n",
+       "      <td>-0.033568</td>\n",
+       "      <td>0.021542</td>\n",
+       "      <td>-0.075791</td>\n",
+       "      <td>0.008654</td>\n",
+       "      <td>True</td>\n",
+       "      <td>998</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>583</th>\n",
+       "      <td>-0.032468</td>\n",
+       "      <td>0.024197</td>\n",
+       "      <td>-0.079893</td>\n",
+       "      <td>0.014957</td>\n",
+       "      <td>True</td>\n",
+       "      <td>583</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>...</th>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "      <td>...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>602</th>\n",
+       "      <td>0.037134</td>\n",
+       "      <td>0.021910</td>\n",
+       "      <td>-0.005809</td>\n",
+       "      <td>0.080077</td>\n",
+       "      <td>True</td>\n",
+       "      <td>602</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>74</th>\n",
+       "      <td>0.040907</td>\n",
+       "      <td>0.025129</td>\n",
+       "      <td>-0.008347</td>\n",
+       "      <td>0.090160</td>\n",
+       "      <td>True</td>\n",
+       "      <td>74</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>732</th>\n",
+       "      <td>0.045107</td>\n",
+       "      <td>0.022044</td>\n",
+       "      <td>0.001900</td>\n",
+       "      <td>0.088314</td>\n",
+       "      <td>False</td>\n",
+       "      <td>732</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>131</th>\n",
+       "      <td>0.046232</td>\n",
+       "      <td>0.021842</td>\n",
+       "      <td>0.003421</td>\n",
+       "      <td>0.089042</td>\n",
+       "      <td>False</td>\n",
+       "      <td>131</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>596</th>\n",
+       "      <td>0.047334</td>\n",
+       "      <td>0.022015</td>\n",
+       "      <td>0.004184</td>\n",
+       "      <td>0.090484</td>\n",
+       "      <td>False</td>\n",
+       "      <td>596</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "<p>100 rows × 6 columns</p>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "           b1     se_b1  ci95_lower  ci95_upper  param_caught   id\n",
+       "765 -0.079756  0.023947   -0.126692   -0.032821         False  765\n",
+       "29  -0.054530  0.021139   -0.095962   -0.013098         False   29\n",
+       "782 -0.039275  0.022142   -0.082672    0.004123          True  782\n",
+       "998 -0.033568  0.021542   -0.075791    0.008654          True  998\n",
+       "583 -0.032468  0.024197   -0.079893    0.014957          True  583\n",
+       "..        ...       ...         ...         ...           ...  ...\n",
+       "602  0.037134  0.021910   -0.005809    0.080077          True  602\n",
+       "74   0.040907  0.025129   -0.008347    0.090160          True   74\n",
+       "732  0.045107  0.022044    0.001900    0.088314         False  732\n",
+       "131  0.046232  0.021842    0.003421    0.089042         False  131\n",
+       "596  0.047334  0.022015    0.004184    0.090484         False  596\n",
+       "\n",
+       "[100 rows x 6 columns]"
+      ]
+     },
+     "execution_count": 18,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "sim_nocluster.sample(100).sort_values('b1')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122594771)>"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(sim_nocluster.sample(100).sort_values('b1'),\n",
+    "                         p.aes(x = 'factor(id)', y = 'b1', \n",
+    "                             ymin = 'ci95_lower', ymax = 'ci95_upper',\n",
+    "                             color = 'param_caught')) +\\\n",
+    "  p.geom_hline(yintercept = sim_params[1], linetype = 'dashed') +\\\n",
+    "  p.geom_pointrange() +\\\n",
+    "  p.labs(x = 'sim ID', y = 'b1', title = 'Randomly Chosen 100 95% CIs') +\\\n",
+    "  p.scale_color_discrete(name = 'True param value', labels = ('missed', 'hit')) +\\\n",
+    "  p.coord_flip()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.04600000000000004"
+      ]
+     },
+     "execution_count": 20,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "1 - sum(sim_nocluster.param_caught)/sim_nocluster.shape[0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions:\n",
+    "\n",
+    "-   What point does the least squares estimate appear to be centered on?\n",
+    "-   Setting the significance level at 5%, we should incorrectly reject the null that $\\beta_1=0$ about 5% of the time in our simulations. About what percent of the time does the 95% confidence intervals contain the true value of $\\beta_1 = 0$?\n",
+    "\n",
+    "### Case 2: Clustered Data\n",
+    "\n",
+    "Now let's resimulate our data with observations that are no longer independent draws in a given cluster of observations, but the true value of $\\beta_1$ still is 0."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Simulation clustered SE\n",
+    "sim_params = [.4, 0] # beta1 = 0: no effect of x on y\n",
+    "sim_nocluster = run_cluster_sim(n_sims=1000, param = sim_params, cluster_robust = False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/stat_bin.py:93: PlotnineWarning: 'stat_bin()' using 'bins = 29'. Pick better value with 'binwidth'.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj4AAAGuCAYAAACQvAxyAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAqoklEQVR4nO3dfXRU9YH/8c/MhJncjAmB8DQIQeVUS6jdrXRXHqQ8GatxQbEsuwv2SCuUg2vdtiu1piDGkmJb9eAKHnHdLadHirqeHk/XlWqAujZhH3rqaVeXYn3gQdzRkCEkw2TCwMz8/rCZHzEBkpk7uTP3+36dkwNzZ+7M5+abe/M5997c60mn02kBAAAYwOt0AAAAgKFC8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGKPE6QBOaGtrczqCPB6PLMtSPB6Xm68h6ff7lUgknI6RV4yle3ijUZU/9ZSiK1cqVV7udJy8MWEsWS/d4+yxrKqqyvn92OPjEK/Xq7KyMnm97h6CQCDgdIS8Yyzdw3vypIZ9//vynjzpdJS8MmIsWS9dw+6xdPdPBAAAwFkoPgAAwBiOn+Pz4osvau/evTp06JBmzJihtWvXSpI++OADbd++XQcOHNCZM2c0adIk3X777frUpz4lSfroo4+0atUqlZaWZt5r7ty5uuOOOxxZDgAAUPgcLz4jR47U0qVL9dvf/lbRaDQzPRaLadq0afr617+uYDCoXbt2qaGhQU899VSvsvP000/L7/c7ER0AABQZxw91zZw5U9OnT1dFRUWv6Zdffrmuv/56VVRUyOfz6S/+4i/U3d2to0ePOpQUAAAUO8f3+AzU22+/rXQ6rVAo1Gv66tWrlU6nNXXqVK1YsUKjR492KCEAACh0RVF8Ojs79cgjj2j58uUKBoOSpIqKCj388MO67LLLFIvFtH37dm3cuFGPPPKIfD5fr/nD4bDC4XDmcSAQ0Pjx44d0GT6pJ+Mns7qNx+Nx/TIylu7h++Ofy/q8XqVdvKxGjCXrpWvYPZYFX3xisZgaGhp01VVX6ZZbbslMtywrc6JzRUWF1qxZo7/6q79SOBzWhAkTer3Htm3b1NDQkHlcX1+vxsbGoVmAC/jkIT43MuUcLMbSBf54nmF5ebk0YoTDYfLL9WP5R6yX7mHXWBZ08enq6tKGDRs0efJkrVq1akDz9HeFztWrV2vRokWZx4FAQO3t7bblzIbP51NFRYU6OzuVTCYdzZJPwWBQsVjM6Rh5xVi6R0k0qnJJ0WhUZxzeRuSTCWPJeukeZ4+lHeXH8eKTTCaVTCaVSqWUSqWUSCTk9XqVSCS0YcMGTZw4UWvWrOkz31tvvaWysjJdfPHF6urq0vbt2xUKhfo9hBUKhXqdG9TW1lYwK0LP8rtVOp129fKdjbEsfp5USpKUTKVcvawmjGUP1kv3sGs5HS8+zz77rJ555pnM45aWFs2fP19XXnml3nrrLR06dEgtLS2Z5zds2KCpU6fqww8/1NNPP60TJ07IsizV1NRo/fr1rj/WCQAAsud48Vm2bJmWLVvW73MLFiw453xz5szRnDlz8hULAAC4kOPX8QEAABgqFB8AAGAMxw91AUAxiUQiisfjWc1rWZaqqqpsTgRgMCg+ADBAkUhENTU1Sv3xr78Gy+v1av/+/ZQfwEEUHwAYoHg8rlQqpdraWlmWNeh5m5qast5bBMAeFB8AGCTLsjK3zwFQXDi5GQAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMSg+AADAGBQfAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxKD4AAMAYFB8AAGAMig8AADAGxQcAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYIwSpwM4we/3KxAIOJrB4/FIkoLBoNLptKNZ8qmkpETl5eVOx8grxtI9PCdOSJLKLEvpfpY1GAzm/BnBYNDx76MRY8l66Rpnj6UdjCw+iURCiUTC0Qw+n09+v1+xWEzJZNLRLPlUXl6uaDTqdIy8YizdoyQel19SVzyuM/0saywWy/kzYrGY499HE8aS9dI9zh5Lv9+f8/txqAsAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMSg+AADAGBQfAABgjBKnAwDAUIpEIorH4/0+V9raqkpJH4bD6k4m+zwfDofzGw5A3lF8ABgjEomopqZGqVSq3+cnSjoi6bovflHvD2kyAEOF4gPAGPF4XKlUSrW1tbIsq8/zo7q6pN27VXvttWorK+vz/PHjx9Xc3DwUUQHkCcUHgHEsy1IwGOwzvafqlJWV9ft8V1dXnpMByDdObgYAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxnD8ys0vvvii9u7dq0OHDmnGjBlau3Zt5rnDhw/rscce06FDhzRu3DitWbNGU6dOzTzf0tKi7du368SJE5oyZYr+7u/+TlVVVU4sBgAAKAKO7/EZOXKkli5dquuuu67X9DNnzmjjxo2aMWOGdu7cqS996UtqbGzUyZMnJUlHjx7VP/zDP+hv//Zv9fTTT2v8+PF66KGHnFgEAABQJBwvPjNnztT06dNVUVHRa/obb7yhU6dOafHixRo2bJjmzZunsWPHat++fZKkX/7yl7rqqqv0p3/6pwoEAlq+fLkOHDigcDjsxGIAAIAi4HjxOZcjR45o0qRJ8nr/f8RLL71UR44ckfTxYbBLL70081x5eblGjx6tw4cPD3lWAABQHBw/x+dc4vF4n7sjB4PBzN2Ru7u7VVZW1uf5eDze573C4XCvPUGBQEDjx4/PQ+qB8/l8vf51K4/H4/plZCyLRyHk9/l8judww1heCOule9g9lgVbfCzLypScHl1dXbIsS5JUWlra5/lYLJZ5/mzbtm1TQ0ND5nF9fb0aGxvzkHrwPnmIz438fr/TEYYEY1n4otGo0xE0fPhwjRgxwukYRT+WA8V66R52jWXBFp/q6mr97Gc/UyqVyhzuOnjwoK6//npJ0qRJk3To0KHM60+ePKm2tjZNmjSpz3utXr1aixYtyjwOBAJqb2/P7wJcgM/nU0VFhTo7O5VMJh3Nkk/BYFCxWMzpGHnFWBaPjo4OpyOoo6PD8e2PG8byQlgv3ePssbSj/DhefJLJpJLJpFKplFKplBKJhLxer6688koNGzZML7zwghYuXKh9+/bpww8/1IwZMyRJc+fO1d13363f/e53+vSnP60dO3boiiuuUCgU6vMZoVCo1/S2traCWRF6lt+t0um0q5fvbIxl4SuE/IXwc+KGsRyoQvh+55NpY2kHx4vPs88+q2eeeSbzuKWlRfPnz9c3vvENrVu3Tlu2bNFPf/pTjR07VvX19SovL5ckTZw4UV//+te1ZcsWtbe3q6amRnfffbdTiwEAAIqA48Vn2bJlWrZsWb/PXXLJJee9Ns8111yja665Jl/RAACAyxTsn7MDAADYjeIDAACMQfEBAADGoPgAAABjUHwAAIAxKD4AAMAYFB8AAGAMig8AADAGxQcAABiD4gMAAIzh+C0rAJgnEokoHo9nNa9lWaqqqrI5EQBTUHwADKlIJKKamhqlUqms5vd6vdq/fz/lB0BWKD4AhlQ8HlcqlVJtba0syxr0vE1NTVnvLQIAig8AR1iWpWAwmNW84XB4SOcD4B4UHwBF49SpU5Kkuro6h5MAKFYUHwBFI5lMSlJWh8kk6fjx42pubrY7FoAiQvEBUHSyPUzW1dWVhzQAignX8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMSg+AADAGEbepNTv9ysQCDiawePxSJKCwaDS6bSjWfKppKRE5eXlTsfIK8ZycLK5uaibBINBx9cJ1kv3MG0s7WBk8UkkEkokEo5m8Pl88vv9isViSiaTjmbJp/LyckWjUadj5BVjOTixWMyGNMUrFos5vk6wXrqHaWPp9/tzfj8OdQEAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMSg+AADAGBQfAABgDCNvUgoATgmHw1nPa1mWqqqqbEwDmIfiAwBD4NSpU5Kkurq6rN/D6/Vq//79lB8gBxQfABgCyWRSklRbWyvLsgY9fzweV1NTk+LxuN3RAKNQfABgCFmWpWAw6HQMwFic3AwAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMQr+JqVLly7t9TiRSOjzn/+81q1bJ0lauXKlTpw4Ia/34w43evRobd26dchzAgCAwlfwxee5557L/D+ZTOr222/XrFmzer3m3nvv1bRp04Y6GgAAKDJFdajr9ddfV3d3t2bOnOl0FAAAUIQKfo/P2fbs2aPZs2crEAj0mr5582al02lVV1fr1ltvVU1NjUMJAQBAISua4tPZ2an//u//1qZNm3pN/9a3vqXJkydL+rgYNTQ06LHHHtOYMWMyrwmHwwqHw5nHgUBA48ePH5rg5+Dz+Xr961Yej8f1y8hYDo7bv0/55vP5cv4esl66B2M5eEVTfF599VWFQiFdccUVvaafvXenrq5Ov/rVr/Sb3/xGN9xwQ2b6tm3b1NDQkHlcX1+vxsbG/IcegIqKCqcj5J3f73c6wpBgLAcmGo3akMRcw4cP14gRI3J+H9ZL92AsB6dois+ePXt07bXXXvB1Xq9X6XS617TVq1dr0aJFmceBQEDt7e22ZxwMn8+niooKdXZ2KplMOpoln4LBoGKxmNMx8oqxHJyOjg4b0piro6Mj5+0X66V7mDaWdpSfoig+7777ro4cOaK5c+f2mn7s2DG1trbq8ssvlyTt3btXb7/9tu68885erwuFQgqFQpnHbW1tBbMiJJPJgsmSD+l02tXLdzbGcmDc/D0aCnb8nLFeuodpY2mHoig+u3fv1uc///k+u3fj8biefPJJhcNhlZSUaOLEiVq/fn2vkgMAANCjKIrP6tWr+51eXV2tRx99dIjTAACAYlVU1/EBAADIBcUHAAAYoygOdQEoLJFIRPF4PKt5z76mFgAMNYoPgEGJRCKqqalRKpVyOgoADBrFB8CgxONxpVIp1dbWyrKsQc9//PhxNTc35yEZAFwYxQdAVizLUjAYHPR8XV1deUgDAAPDyc0AAMAYFB8AAGAMig8AADAGxQcAABiD4gMAAIxB8QEAAMbIqvjMnz9fBw4c6Pe5P/zhD5o/f35OoQAAAPIhq+Lz6quvqrOzs9/nOjs79dprr+UUCgAAIB+yPtTl8Xj6nb5v3z6NGTMm60AAAAD5MuArN2/atEmbNm2S9HHpmTdvnrze3r3p1KlTOnPmjO644w57UwIAANhgwMVn5syZ+vu//3ul02k98MAD+pu/+RtNmDCh12v8fr+mTJmihQsX2h4UAAAgVwMuPnPmzNGcOXMkfbzHZ9WqVRo/fnzeggEAANgtq5uUbtiwwe4cAAAAeZdV8UmlUnrqqaf0/PPP6+jRo+ru7u71vMfj0bvvvmtLQAAAALtkVXzuuecePfzww5o1a5Zmz54tv99vdy4AAADbZVV8duzYofvvv1/33Xef3XkAAADyJqvr+HR3d2vWrFl2ZwEAAMirrIrP8uXL9a//+q92ZwEAAMirrA51TZ8+XevWrdNHH32k2tpaVVZW9nnNLbfckms2AAAAW2VVfL785S9Lkg4fPqxnn322z/Mej0fJZDK3ZAAAADbLqvgcPHjQ7hwAAAB5l1XxmTRpkt05AAAA8i6r4nPkyJELvqa6ujqbtwYAAMibrIrPJZdcIo/Hc97XFPI5Pn6/X4FAwNEMPd+/YDCodDrtaJZ8KikpUXl5udMx8sq0sQwGg05HMVowGMx5nWK9dA/TxtIOWRWff/mXf+kz7fjx43rllVf061//Wo2NjTkHy6dEIqFEIuFoBp/PJ7/fr1gsVtAlMVfl5eWKRqNOx8gr08YyFos5HcVosVgs53WK9dI9TBtLO+4UkVXx+dKXvtTv9FWrVumb3/ymWlpatHz58pyCAQAA2C2rCxiez4033qhnnnnG7rcFAADIme3FZ9++fSotLbX7bQEAAHKW1aGuu+66q8+0RCKh3//+92pubtbdd9+dczAAAAC7ZVV8+rtPV2lpqSZMmKDHH39cK1euzDkYAACA3bhyMwAAMEbO5/ik02lFo1FXXycBAAC4Q1Z7fCTp3//939XQ0KB9+/bp9OnTGjZsmGbNmqX7779fs2fPtjMjAOCPwuFwVvNZlqWqqiqb0wDFJ6vi09TUpLq6Ol1++eW69957NW7cOIXDYT3//PNasGCBXnrpJV177bV2ZwUAY506dUqSVFdXl9X8Xq9X+/fvp/zAeFkVn3Xr1qmurk4vvPBCr1tXbNiwQTfffLPWrVtH8QEAG/Vcfbi2tlaWZQ1q3ng8rqamJsXj8XxEA4pKVsXnjTfeUENDQ5/7dXk8Hq1Zs0a33HKLLeEAAL1ZlsX90oAcZHVy80UXXaQPPvig3+eOHj2qiy66KKdQAAAA+ZBV8Vm0aJG+853v6OWXX+41/ZVXXtF3v/td3XTTTbaEAwAAsFNWh7p+9KMf6Y033tANN9ygiooKjR07Vh999JGi0aj+7M/+TD/60Y/szgkAAJCzrIrPiBEj9B//8R968cUX1dzcrPb2do0cOVLXXHONbrzxRnm9tt8CDAAAIGdZFZ89e/boyJEj+spXvqJFixb1em779u2aNGmS5s2bZ0tAAIA9eq4BFAwGFYvFBjUv1wGCW2T95+znOo/n2LFj+sd//Ee1tLTkFAwAYI9crwEkcR0guEdWxed///d/9b3vfa/f56666io1NjbmFAoAYJ9crgEkcR0guEtWxcfj8aijo6Pf59rb2zMrGQCgcHANICDLP2e/+uqrtXXr1j43Jk2n03r88cd19dVX2xIOAADATlnt8WloaNC8efP02c9+VitWrFAoFNL//d//6Sc/+Yn+8Ic/6NVXX7U5JgAAQO6yKj4zZszQnj179O1vf1v33HOPUqmUvF5vZvr06dPtzgkAAJCzrIqPJM2aNUstLS2Kx+Nqb29XZWWlysrK7MwGAABgq6yLTw/LsrL6KwEAAIChxiWWAQCAMSg+AADAGBQfAABgDIoPAAAwRs4nN+fb5s2b9dprr6mk5P9H3bp1q0aPHi1JOnz4sB577DEdOnRI48aN05o1azR16lSn4gIAgAJW8MVHkm666SbddtttfaafOXNGGzdu1PXXX69NmzapublZjY2NevLJJ3XRRRc5kBQAABSyoj7U9cYbb+jUqVNavHixhg0bpnnz5mns2LHat2+f09EAAEABKoo9Pi+//LJefvlljRo1SgsXLlRtba0k6ciRI5o0aZK83v/f3y699FIdOXLEqagAAKCAFXzxWbhwob761a8qGAxq//79evDBBxUMBjVz5kzF4/E+dxoOBoPq6urqNS0cDiscDmceBwIBjR8/fkjyn4vP5+v1r1t5PB7XL6NpY+n25cS5FdP4m7ZeupndY1nwxWfy5MmZ/1955ZW68cYb1dLSopkzZ8qyrD4lp6urq8+VpLdt26aGhobM4/r6ejU2NuY3+ABVVFQ4HSHv/H6/0xGGhCljGY1GnY4BhwwfPlwjRoxwOsagmLJemsCusSz44vNJHo9H6XRaklRdXa2f/exnmZukStLBgwd1/fXX95pn9erVWrRoUeZxIBBQe3v70IXuh8/nU0VFhTo7O5VMJh3Nkk/BYFCxWMzpGHll2lh2dHQ4HQUO6ejocHzbOVCmrZdudvZY2lF+Cr74NDc366qrrlJpaakOHDigf/u3f9PXvvY1SR/vARo2bJheeOEFLVy4UPv27dOHH36oGTNm9HqPUCikUCiUedzW1lYwK0IymSyYLPmQTqddvXxnM2Us3byMOL9iHP9izDwYpm1j7VDwxefFF1/U1q1blUqlNGrUKC1fvlxf+MIXJEklJSVat26dtmzZop/+9KcaO3as6uvrVV5e7nBqAABQiAq++Dz44IPnff6SSy7RQw89NERpgMIRiUQUj8eznr+7u1ulpaUDfn3PLvWz/1AAAIpNwRcfAH1FIhHV1NQolUo5HQUAigrFByhC8XhcqVRKtbW1ff6KcSCOHz+u5ubmrObvmRcAihHFByhilmX1uZbVQPRcBiKb+T95CQkAKCZFfcsKAACAwaD4AAAAY1B8AACAMSg+AADAGBQfAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxKD4AAMAYFB8AAGAMig8AADAGxQcAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY5Q4HcAJfr9fgUDA0Qwej0eSFAwGlU6nHc2STyUlJSovL3c6Rt60tbUpHo8rGo0qkUgMeizLyso0atSoQX9uMBgc9DxAroLBYNGsz2xj3ePssbSDkcUnkUgokUg4msHn88nv9ysWiymZTDqaJZ/Ky8sVjUadjpEXkUhENTU1SqVSWb+H1+vV/v37VVVVNaj5YrFY1p8JZCsWixXN+sw21j3OHku/35/z+xlZfAA7xONxpVIp1dbWyrKsrOZvampSPB7PQzoAQH8oPkCOLMvi0BMAFAlObgYAAMag+AAAAGNwqAtwWDgcHpJ5gFxl+3NnWdagT+AH8oXiAzjk1KlTkqS6ujqHkwDnl+vParZ/vQjkA8UHcEjPn9hm81dhx48fV3Nzcz5iAX3k8rPKXy+i0FB8AIdl81dhXV1deUoDnBt/wQg34ORmAABgDIoPAAAwBsUHAAAYg+IDAACMwcnNAIC8y+XaU1wHCHai+AAA8saO61VxHSDYieIDAMibXK4BJHEdINiP4gMAyDuuAYRCwcnNAADAGBQfAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxKD4AAMAYFB8AAGAMblkBACh4g727u8/nUzQaVSKRUGVlZX5CoSgVfPE5ffq0nnjiCf3ud79TNBrVqFGj9Jd/+ZeaO3euJGnlypU6ceKEvN6Pd16NHj1aW7dudTAxAMAuud7dnTu745MKvvgkk0mNHDlSGzdu1JgxY3TgwAE98MADGjdunD796U9Lku69915NmzbN4aQAALvlcnd37uyO/hR88SktLdXy5cszj2tqajRlyhT9/ve/zxQfAIC7cXd32KXgi88ndXd365133tHChQsz0zZv3qx0Oq3q6mrdeuutqqmpcTAhAAAoVEVVfNLptB599FF96lOf0uc+9zlJ0re+9S1NnjxZkrRnzx41NDToscce05gxYzLzhcPhXifGBQIBjR8/fmjDf4LP5+v1r1t5PB7XLqNblwtwG5/P59r11c3b2B52/74smuKTTqf1+OOPKxKJ6IEHHpDH45GkXnt36urq9Ktf/Uq/+c1vdMMNN2Smb9u2TQ0NDZnH9fX1amxsHLrw51FRUeF0hLzz+/1OR8iLaDTqdAQAAzB8+HCNGDHC6Rh549Zt7CfZ9fuyKIpPOp3WE088offee0/f+973VFpaes7Xer1epdPpXtNWr16tRYsWZR4HAgG1t7fnLe9A+Hw+VVRUqLOzM3PynhsFg0HFYjGnY+RFR0eH0xEADEBHR4fj2/x8cfM2tsfZvy/tKD9FUXy2bdumt956Sxs3blRZWVlm+rFjx9Ta2qrLL79ckrR37169/fbbuvPOO3vNHwqFFAqFMo/b2toKpmwkk8mCyZIP6XTatcvn1uUC3MbN21k3b2M/ya7lLPji09raqpdeeknDhg3TV7/61cz0JUuWaPr06XryyScVDodVUlKiiRMnav369b1KDgAAQI+CLz5jxozRz3/+83M+/+ijjw5hGgAAUMy4VxcAADAGxQcAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGIPiAwAAjEHxAQAAxqD4AAAAY1B8AACAMSg+AADAGBQfAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxSpwOADgtEokoHo8Per5wOJyHNACAfKL4wGiRSEQ1NTVKpVJORwEADAGKD4wWj8eVSqVUW1sry7IGNe/x48fV3Nycp2QAgHyg+ACSLMtSMBgc1DxdXV15SgMAyBdObgYAAMag+AAAAGNQfAAAgDEoPgAAwBhGntzs9/sVCAQczeDxeCRJwWBQ6XTa0Sz5VFJSovLycqdjnNNgT2gGUHyCwWBBb4dyUejbWDuc/fvSDkYWn0QioUQi4WgGn88nv9+vWCymZDLpaJZ8Ki8vVzQadTrGOcViMacjAMizWCxW0NuhXBT6NtYOZ/++9Pv9Ob8fh7oAAIAxKD4AAMAYFB8AAGAMig8AADCGkSc3AwAwEJFIRPF4PKt5LctSVVWVzYmQK4oPAAD9iEQiqqmpUSqVymp+r9er/fv3U34KDMUHAIB+xONxpVIp1dbWyrKsQc/b1NSU9d4i5A/FBwCA87Asi4uduggnNwMAAGOwxwdFL5eTD8PhsM1pAACFjOKDopbryYcAALNQfFDUcjn5UJKOHz+u5ubmPCQDABQiig9cIduTD7u6uvKQBkAhyfaQNofC3YniAwBwpVOnTkmS6urqHE6CQkLxAQC4UjKZlCQOhaMXig8AwNU4FI6zcR0fAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxuI4PAAB5ksttLyzLUlVVlY1pIFF8AACwnR23y/B6vdq/fz/lx2YUHwAAbJbr7TLi8biampoUj8ftjmY8ig8kSZFIJKcVrLu7W6WlpX2mB4NBxWKxC87PLl0AbpTt7TLskMt23c3b5KIvPidPntTWrVv1+uuvy7IsLV26lDvxDlIkElFNTY1SqZRjGdilCwD2yXW77uZtctEXn23btimZTOrHP/6xwuGw7rvvPk2YMEGf/exnHckz0Ibt8/kUjUbV0dGR2SWaa8POtt2Hw2GlUqmc72Cc6y7d9957b9D5czlxEAAK3YW2cefaq57Ldj2XbXKPQt5jVNTFp7u7Wy0tLdq8ebPKyso0efJkzZ8/X7t373ak+DjZsO3Ya5PrHYyznd+OkwABwE3s2i5ms112+4nZRV18PvjgA0lSdXV1Ztpll12mF154wZE88Xg854adbbvO5bN79tg4JZeTAJ3ODgD5kOvJ0blsG91+YnZRF5/u7u4+gxIMBvt8s8PhcK/dhYFAQOPHj7c9j8/ny/k9Wltbs3qf1tbWnD872x/S7u5uW+bPhRPZ7VpuPrtwPtv6497Lrq4u9XdKfiFn57Pd+dm5yiV7rnw+ny2/F3vew473kiSli9g777yTXrx4ca9pe/fuTd911129pm3YsCEtKfNVX1+flzzHjh1Le73eXp81lF8ej8exz+aLLzd8TZTS6T/+63QWvvgq5i+v15s+duxYXn7X5qqo9/hcfPHFkqT3339fEydOlCQdPHhQkyZN6vW61atXa9GiRZnHgUBA7e3ttufx+Xx66623BtSwvV6vysvLFY1GM+flnDlzRiUl2Q9JLvPH4/GsdmleaH7Lsgb0/cjl8/OVfaDzBoPBPmM5VJ89lMt99lg6/T3P12cHPvpIqq3V7qYmnRo71vbPL5Tv20DXy3x89lDN37ONPXbsmPx+/5B+th3zDnT+842lk9kty5LP57Pld63P51NFRYU6OztVUVGR8/sVdfEpLS3VrFmztGPHDt1111366KOPtGfPHn3729/u9bpQKKRQKJR53NbWljmGabfKykpVVlZe8HU+n08jRoxQe3t73rIUgp5C4GaMpXv0bBDHjB2rM2dtM9zGhLHsWS/Ly8tZLx1i9/fdrvcr6uIjfbw3Z8uWLVqxYoXKysq0fPly/cmf/InTsQAAQAEq+uJz0UUX6Tvf+Y7TMQAAQBHwOh0AAABgqFB8AACAMSg+AADAGBQfAABgDIoPAAAwBsUHAAAYg+IDAACMQfEBAADGoPgAAABjUHwAAIAxKD4AAMAYFB8AAGAMTzqdTjsdwkThcFjbtm3T6tWrFQqFnI6DHDCW7sFYugdj6R52jyV7fBwSDofV0NCgcDjsdBTkiLF0D8bSPRhL97B7LCk+AADAGBQfAABgDIqPQ0KhkDZs2MCxZxdgLN2DsXQPxtI97B5LTm4GAADGYI8PAAAwBsUHAAAYo8TpACZ5+umntWvXLqVSKc2ePVtf+9rXVFLSdwhOnDihp556Sm+++abi8bguvvhiffnLX9bnPvc5B1KjPwMdy57X/td//Zfef/99LV68WLfddtsQp8VAnDx5Ulu3btXrr78uy7K0dOlS1dXVOR0L5zHQMTt9+rQefvhhvfPOO2ptbdWGDRs0bdo0BxLjXAY6lgcOHNDOnTv1zjvvSJKuuOIKrVy5UuPHjx/wZ7HHZ4i88soreu211/TII4/oiSee0Hvvvafnnnuu39d2d3dr8uTJevjhh7Vz504tWbJEmzZtUmtr6xCnRn8GM5bSxyfmrVixQn/+538+hCkxWNu2bVMymdSPf/xjrV+/Xjt27ND//M//OB0L5zGYMZsyZYq++c1vatSoUUOcEgMx0LGMxWK69tpr9eSTT2r79u2qrq7Wxo0bB/VZFJ8hsnv3bt18880aO3ashg8frqVLl2r37t39vnbcuHFavHixqqqq5PV6NXPmTI0ePTrTcOGswYylJC1YsEDTpk1TWVnZEKbEYHR3d6ulpUW33nqrysrKNHnyZM2fP/+84wpnDWbMhg0bpptuuklTp06V18uvvUIzmLGcNm2aZs+erWAwqGHDhunmm2/W0aNH1dnZOeDP4ydgiBw5ckSXXHJJ5vGll16qtrY2xWKxC84biUQUDodVXV2dx4QYqFzGEoXpgw8+kKRe69hll12mw4cPOxUJF8CYuUcuY/nmm29qxIgRqqioGPDncY6PDZLJ5Hmf9/l86u7uVjAYzEzr+X88Hu81/ZMSiYR++MMf6rrrrtOECRPsCYxzyudYonB1d3fLsqxe04LBoOLxuEOJcCGMmXtkO5Yffvhh5h5eg0HxscH69ev15ptv9vtcZWWlfvKTn6i0tFRdXV2Z6T3//+Rgn+306dN68MEHVVlZqVWrVtkbGv3K11iisJWWlvbZyMZiMca0gDFm7pHNWLa1tem+++7TkiVLdM011wzq8yg+Nvj+979/wddUV1fr4MGDmjJliiTp4MGDGjVq1Dn3EJw+fVo/+MEP5PV6tXbtWvl8Plszo3/5GEsUvosvvliS9P7772vixImSPh7XSZMmORkL58GYucdgxzISiei73/2urrvuOt10002D/jzO8RkiCxYs0M9//nO1traqs7NTzz77rK699tp+X3vmzBn98Ic/1OnTp3XPPfec88+k4YzBjKX08XgmEgmlUimlUiklEokLHlLD0CotLdWsWbO0Y8cOdXV16eDBg9qzZ48WLFjgdDScw2DH7PTp00okEkqn00omk5l1Es4bzFhGIhHV19dr7ty5WrJkSVafxy0rhkg6ndaOHTu0a9cuJZNJfeELX+h17Zf7779fNTU1Wrp0qd58803V19fL7/f3+guEO+64Q3PnznVoCdBjMGMpSZs3b9bevXt7vcdf//Vfa9myZUOeHed28uRJbdmyRa+//rrKysq4jk8RON+YLV26VBs2bNDUqVMlSStXruxzSZDGxkZdeeWVQ54bfQ10LHfu3KmdO3eqtLS01/xbt27V6NGjB/RZFB8AAGAMDnUBAABjUHwAAIAxKD4AAMAYFB8AAGAMig8AADAGxQcAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAUlBUrVugzn/nMeV/T1NSkZcuWafLkyfJ4PLrzzjuHKB2AYkfxAVB0du3apd/+9reaM2eOKisrnY4DoIhQfAAUnYceekj79+/XP//zP2v48OFOxwFQRCg+AArSrl279JnPfEalpaWaNm2a/vM//zPznNfLpgtAdth6ACg44XBYd9xxh9auXavnnntOgUBAX/ziF9Xa2up0NABFjuIDoOAcP35c//RP/6TbbrtNixYt0ksvvaRUKqXNmzc7HQ1AkaP4ACg4w4cP1/z58zOPKysrNX/+/F6HuwAgGxQfAAVn9OjRfaaNGTNG4XDYgTQA3ITiA6DgHDt2rM+01tZWhUIhB9IAcBOKD4CC09HRob179/Z5fPXVVzuYCoAblDgdAAA+aeTIkbr99tvV0NCgyspKPfjgg5Kkb3zjG5Kkw4cP69e//rUkqaurS++++66ef/55SdKSJUscyQygOFB8ABScUCikH/zgB1q7dq3effddTZ06VS+//LLGjh0rSfrlL3+pr3zlK5nX/+IXv9AvfvELSVI6nXYkM4Di4EmzlQAAAIbgHB8AAGAMig8AADAGxQcAABiD4gMAAIxB8QEAAMag+AAAAGNQfAAAgDEoPgAAwBgUHwAAYAyKDwAAMAbFBwAAGOP/AXH7lqlUuqcQAAAAAElFTkSuQmCC\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122786274)>"
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(sim_nocluster, p.aes('b1')) +\\\n",
+    "  p.geom_histogram(color = 'black') +\\\n",
+    "  p.geom_vline(xintercept = sim_params[1], color = 'red')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8782122670141)>"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(sim_nocluster.sample(100).sort_values('b1'),\n",
+    "                         p.aes(x = 'factor(id)', y = 'b1', \n",
+    "                             ymin = 'ci95_lower', ymax = 'ci95_upper',\n",
+    "                             color = 'param_caught')) +\\\n",
+    "  p.geom_hline(yintercept = sim_params[1], linetype = 'dashed') +\\\n",
+    "  p.geom_pointrange() +\\\n",
+    "  p.labs(x = 'sim ID', y = 'b1', title = 'Randomly Chosen 100 95% CIs') +\\\n",
+    "  p.scale_color_discrete(name = 'True param value', labels = ('missed', 'hit')) +\\\n",
+    "  p.coord_flip()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.398"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "1 - sum(sim_nocluster.param_caught)/sim_nocluster.shape[0]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions:\n",
+    "\n",
+    "-   When the errors are clustered, does the distribution of $\\hat{\\beta}_1$ estimates get wider or narrower?\n",
+    "-   When the errors are clustered, do we incorrectly reject the null more or less frequently?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
diff --git a/Python/README.md b/Python/README.md
new file mode 100644
index 0000000..4397aab
--- /dev/null
+++ b/Python/README.md
@@ -0,0 +1,9 @@
+# Notes on Python Implementation
+
+Code outputs have been verified and match R output except in the following cases:
+
+- **Differences_in_Differences.ipynb:** The design matrix in the Cunningham and Cornwell (2013) example is rank deficient. lm and lm_robust have a convergence issues. The problem seems to be with the fip variable. The statsmodels algorithm is more robust to rank deficiency resulting is different results
+- **Matching_and_Subclassification.ipynb:** Python does not have an implementation of nearest neighbor matching in python. I may be possible to recreate one use standard KNN tools
+- **Regression_Discontinuity.ipynb:** The smoothing and density section is missing
+
+The majority of the models use the statmodels package. While I tried to limit my usage of R code through rpy2 I use it once for synthetic control matching.
diff --git a/Python/Regression_Discontinuity.ipynb b/Python/Regression_Discontinuity.ipynb
new file mode 100644
index 0000000..49eb8ab
--- /dev/null
+++ b/Python/Regression_Discontinuity.ipynb
@@ -0,0 +1,927 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Regression Discontinuity** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import plotnine as p\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "from stargazer.stargazer import Stargazer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def read_data(file):\n",
+    "    full_path = \"https://raw.github.com/scunning1975/mixtape/master/\" + file\n",
+    "    \n",
+    "    return pd.read_stata(full_path)\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## RDD Simulation\n",
+    "\n",
+    "Generate a simple Regression Discontinuity, before and after the treatment is given. "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\"Counterfactual Potential Outcomes\n"
+     ]
+    }
+   ],
+   "source": [
+    "dat = pd.DataFrame({'x': np.random.normal(50, 25, 1000)})\n",
+    "dat.loc[dat.x<0, 'x'] = 0\n",
+    "dat = dat[dat.x<100]\n",
+    "dat['D'] = 0\n",
+    "dat.loc[dat.x>50, 'D'] = 1\n",
+    "dat['y1'] = 25 + 0*dat.D + 1.5 * dat.x + np.random.normal(0, 20, dat.shape[0])\n",
+    "dat['y2'] = 25 + 40*dat.D + 1.5 * dat.x + np.random.normal(0, 20, dat.shape[0])\n",
+    "print('\"Counterfactual Potential Outcomes')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781696931398)>"
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(dat, p.aes(x='x', y='y1', color = 'factor(D)')) +\\\n",
+    "    p.geom_point(alpha = 0.5) +\\\n",
+    "    p.geom_vline(xintercept = 50, colour = \"grey\") +\\\n",
+    "    p.stat_smooth(method = \"lm\", se = 'F') +\\\n",
+    "    p.labs(x = \"Test score (X)\", y = \"Potential Outcome (Y)\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\"Counterfactual Potential Outcomes after Treatment\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781697517930)>"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "print('\"Counterfactual Potential Outcomes after Treatment')\n",
+    "p.ggplot(dat, p.aes(x='x', y='y2', color = 'factor(D)')) +\\\n",
+    "    p.geom_point(alpha = 0.5) +\\\n",
+    "    p.geom_vline(xintercept = 50, colour = \"grey\") +\\\n",
+    "    p.stat_smooth(method = \"lm\", se = 'F') +\\\n",
+    "    p.labs(x = \"Test score (X)\", y = \"Potential Outcome (Y)\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "dat = pd.DataFrame({'x': np.random.normal(100, 50, 1000)})\n",
+    "dat.loc[dat.x<0, 'x'] = 0\n",
+    "dat['x2'] = dat['x']**2\n",
+    "dat['x3'] = dat['x']**3\n",
+    "dat['D'] = 0\n",
+    "dat.loc[dat.x>140, 'D'] = 1\n",
+    "\n",
+    "dat['y3'] = 10000 + 0*dat.D - 100 * dat.x + dat.x2 + np.random.normal(0, 1000, 1000)\n",
+    "dat = dat[dat.x < 280]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781697523550)>"
+      ]
+     },
+     "execution_count": 7,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Linear Model for conditional expectation\n",
+    "p.ggplot(dat, p.aes(x='x', y='y3', color = 'factor(D)')) +\\\n",
+    "    p.geom_point(alpha = 0.2) +\\\n",
+    "    p.geom_vline(xintercept = 140, colour = \"grey\") +\\\n",
+    "    p.stat_smooth(method = \"lm\", se = 'F') +\\\n",
+    "    p.labs(x = \"Test score (X)\", y = \"Potential Outcome (Y)\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/smoothers.py:310: PlotnineWarning: Confidence intervals are not yet implementedfor lowess smoothings.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781697521050)>"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Linear Model for conditional expectation\n",
+    "p.ggplot(dat, p.aes(x='x', y='y3', color = 'factor(D)')) +\\\n",
+    "    p.geom_point(alpha = 0.2) +\\\n",
+    "    p.geom_vline(xintercept = 140, colour = \"grey\") +\\\n",
+    "    p.stat_smooth(method = \"lowess\", se = 'F') +\\\n",
+    "    p.labs(x = \"Test score (X)\", y = \"Potential Outcome (Y)\")\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "#### Questions\n",
+    "- If you estimated the treatment effect using the linear model, would your treatment effect estimate be near the true treatment effect? \n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<table class=\"simpletable\">\n",
+       "<caption>OLS Regression Results</caption>\n",
+       "<tr>\n",
+       "  <th>Dep. Variable:</th>           <td>y3</td>        <th>  R-squared:         </th> <td>   0.978</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.978</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   6445.</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Date:</th>             <td>Sun, 07 Mar 2021</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Time:</th>                 <td>13:34:36</td>     <th>  Log-Likelihood:    </th> <td> -8302.1</td> \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>No. Observations:</th>      <td>  1000</td>      <th>  AIC:               </th> <td>1.662e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Residuals:</th>          <td>   992</td>      <th>  BIC:               </th> <td>1.666e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Df Model:</th>              <td>     7</td>      <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Intercept</th> <td> 1.036e+04</td> <td>  155.486</td> <td>   66.635</td> <td> 0.000</td> <td> 1.01e+04</td> <td> 1.07e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>D</th>         <td>-1572.2293</td> <td> 1.54e+04</td> <td>   -0.102</td> <td> 0.919</td> <td>-3.17e+04</td> <td> 2.86e+04</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>x</th>         <td> -119.1308</td> <td>    8.641</td> <td>  -13.787</td> <td> 0.000</td> <td> -136.088</td> <td> -102.174</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>x2</th>        <td>    1.2875</td> <td>    0.141</td> <td>    9.152</td> <td> 0.000</td> <td>    1.011</td> <td>    1.564</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>x3</th>        <td>   -0.0013</td> <td>    0.001</td> <td>   -1.951</td> <td> 0.051</td> <td>   -0.003</td> <td> 7.54e-06</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>D:x</th>       <td>   37.1505</td> <td>  246.124</td> <td>    0.151</td> <td> 0.880</td> <td> -445.832</td> <td>  520.133</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>D:x2</th>      <td>   -0.3930</td> <td>    1.298</td> <td>   -0.303</td> <td> 0.762</td> <td>   -2.941</td> <td>    2.155</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>D:x3</th>      <td>    0.0015</td> <td>    0.002</td> <td>    0.650</td> <td> 0.516</td> <td>   -0.003</td> <td>    0.006</td>\n",
+       "</tr>\n",
+       "</table>\n",
+       "<table class=\"simpletable\">\n",
+       "<tr>\n",
+       "  <th>Omnibus:</th>       <td> 2.713</td> <th>  Durbin-Watson:     </th> <td>   1.998</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Prob(Omnibus):</th> <td> 0.258</td> <th>  Jarque-Bera (JB):  </th> <td>   2.577</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Skew:</th>          <td> 0.101</td> <th>  Prob(JB):          </th> <td>   0.276</td>\n",
+       "</tr>\n",
+       "<tr>\n",
+       "  <th>Kurtosis:</th>      <td> 3.144</td> <th>  Cond. No.          </th> <td>2.00e+09</td>\n",
+       "</tr>\n",
+       "</table><br/><br/>Notes:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large,  2e+09. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
+      ],
+      "text/plain": [
+       "<class 'statsmodels.iolib.summary.Summary'>\n",
+       "\"\"\"\n",
+       "                            OLS Regression Results                            \n",
+       "==============================================================================\n",
+       "Dep. Variable:                     y3   R-squared:                       0.978\n",
+       "Model:                            OLS   Adj. R-squared:                  0.978\n",
+       "Method:                 Least Squares   F-statistic:                     6445.\n",
+       "Date:                Sun, 07 Mar 2021   Prob (F-statistic):               0.00\n",
+       "Time:                        13:34:36   Log-Likelihood:                -8302.1\n",
+       "No. Observations:                1000   AIC:                         1.662e+04\n",
+       "Df Residuals:                     992   BIC:                         1.666e+04\n",
+       "Df Model:                           7                                         \n",
+       "Covariance Type:            nonrobust                                         \n",
+       "==============================================================================\n",
+       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
+       "------------------------------------------------------------------------------\n",
+       "Intercept   1.036e+04    155.486     66.635      0.000    1.01e+04    1.07e+04\n",
+       "D          -1572.2293   1.54e+04     -0.102      0.919   -3.17e+04    2.86e+04\n",
+       "x           -119.1308      8.641    -13.787      0.000    -136.088    -102.174\n",
+       "x2             1.2875      0.141      9.152      0.000       1.011       1.564\n",
+       "x3            -0.0013      0.001     -1.951      0.051      -0.003    7.54e-06\n",
+       "D:x           37.1505    246.124      0.151      0.880    -445.832     520.133\n",
+       "D:x2          -0.3930      1.298     -0.303      0.762      -2.941       2.155\n",
+       "D:x3           0.0015      0.002      0.650      0.516      -0.003       0.006\n",
+       "==============================================================================\n",
+       "Omnibus:                        2.713   Durbin-Watson:                   1.998\n",
+       "Prob(Omnibus):                  0.258   Jarque-Bera (JB):                2.577\n",
+       "Skew:                           0.101   Prob(JB):                        0.276\n",
+       "Kurtosis:                       3.144   Cond. No.                     2.00e+09\n",
+       "==============================================================================\n",
+       "\n",
+       "Notes:\n",
+       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
+       "[2] The condition number is large,  2e+09. This might indicate that there are\n",
+       "strong multicollinearity or other numerical problems.\n",
+       "\"\"\""
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "np.random.seed(12282020)\n",
+    "\n",
+    "# Fully interacted regression\n",
+    "all_columns = \"+\".join(dat.columns.difference([\"D\", 'y3']))\n",
+    "formula = 'y3 ~ D * ({})'.format(all_columns)\n",
+    "\n",
+    "regression = sm.OLS.from_formula(formula, data = dat).fit()\n",
+    "regression.summary()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "collapsed": true
+   },
+   "source": [
+    "#### Question\n",
+    "- What did you estimate as the treatment effect using a third order polynomial of the running variable? Is it statistically significantly different from zero?\n",
+    "- Does the estimated treatment effect seem correct from the graph?\n",
+    "\n",
+    "## The Close Election Design\n",
+    "\n",
+    "Lets load the data from Lee et. al. (2004):"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def lm_robust(formula, data):\n",
+    "    regression = sm.OLS.from_formula(formula, data = data)\n",
+    "    regression = regression.fit(cov_type=\"cluster\",cov_kwds={\"groups\":data['id']})\n",
+    "    return regression"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "lmb_data = read_data(\"lmb-data.dta\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "lmb_data['demvoteshare_c'] = lmb_data['demvoteshare'] - 0.5\n",
+    "lmb_subset = lmb_data[lmb_data.lagdemvoteshare.between(.48, .52)]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Original results based on ADA Scores -- Close Elections Sample\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>31.196<sup>***</sup></td><td>18.747<sup>***</sup></td><td>0.242<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1.334)</td><td>(0.843)</td><td>(0.020)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>47.706<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(1.356)</td><td></td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>21.284<sup>***</sup></td><td></td><td>0.484<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1.951)</td><td></td><td>(0.029)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>915</td><td>915</td><td>915</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.115</td><td>0.578</td><td>0.235</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.114</td><td>0.578</td><td>0.234</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>29.522 (df=913)</td><td>20.380 (df=913)</td><td>0.438 (df=913)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>118.982<sup>***</sup> (df=1; 913)</td><td>1237.691<sup>***</sup> (df=1; 913)</td><td>280.233<sup>***</sup> (df=1; 913)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca62c60e50>"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm_1 = lm_robust('score ~ lagdemocrat', data = lmb_subset)\n",
+    "lm_2 = lm_robust('score ~ democrat', data = lmb_subset)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat', data = lmb_subset)\n",
+    "print(\"Original results based on ADA Scores -- Close Elections Sample\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### using all data (note data used is lmb_data, not lmb_subset)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Original results based on ADA Scores -- Full Sample\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>23.539<sup>***</sup></td><td>17.576<sup>***</sup></td><td>0.120<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.337)</td><td>(0.263)</td><td>(0.004)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>40.763<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.418)</td><td></td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>31.506<sup>***</sup></td><td></td><td>0.818<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.484)</td><td></td><td>(0.005)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>13,588</td><td>13,588</td><td>13,588</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.227</td><td>0.376</td><td>0.676</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.227</td><td>0.376</td><td>0.676</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>28.694 (df=13586)</td><td>25.785 (df=13586)</td><td>0.279 (df=13586)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>4243.294<sup>***</sup> (df=1; 13586)</td><td>9502.258<sup>***</sup> (df=1; 13586)</td><td>25739.792<sup>***</sup> (df=1; 13586)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca62bdf520>"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm_1 = lm_robust('score ~ lagdemocrat', data = lmb_data)\n",
+    "lm_2 = lm_robust('score ~ democrat', data = lmb_data)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat', data = lmb_data)\n",
+    "print(\"Original results based on ADA Scores -- Full Sample\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# drop missing values\n",
+    "lmb_data = lmb_data[~pd.isnull(lmb_data.demvoteshare_c)]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Results based on ADA Scores -- Full Sample\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>22.883<sup>***</sup></td><td>11.034<sup>***</sup></td><td>0.212<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.443)</td><td>(0.336)</td><td>(0.005)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>58.502<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.656)</td><td></td></tr><tr><td style=\"text-align:left\">demvoteshare_c</td><td>-5.626<sup>***</sup></td><td>-48.938<sup>***</sup></td><td>0.773<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1.898)</td><td>(1.641)</td><td>(0.019)</td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>33.451<sup>***</sup></td><td></td><td>0.552<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.848)</td><td></td><td>(0.010)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>13,577</td><td>13,577</td><td>13,577</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.227</td><td>0.424</td><td>0.735</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.227</td><td>0.424</td><td>0.735</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>28.686 (df=13574)</td><td>24.764 (df=13574)</td><td>0.252 (df=13574)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>2115.714<sup>***</sup> (df=2; 13574)</td><td>6192.133<sup>***</sup> (df=2; 13574)</td><td>48967.170<sup>***</sup> (df=2; 13574)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca62c508e0>"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\n",
+    "lm_1 = lm_robust('score ~ lagdemocrat + demvoteshare_c', data = lmb_data)\n",
+    "lm_2 = lm_robust('score ~ democrat + demvoteshare_c', data = lmb_data)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat + demvoteshare_c', data = lmb_data)\n",
+    "print(\"Results based on ADA Scores -- Full Sample\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Original results based on ADA Scores -- Full Sample with linear interactions\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>31.435<sup>***</sup></td><td>16.816<sup>***</sup></td><td>0.287<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.541)</td><td>(0.418)</td><td>(0.008)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>55.431<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.637)</td><td></td></tr><tr><td style=\"text-align:left\">democrat:demvoteshare_c</td><td></td><td>-55.152<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(3.218)</td><td></td></tr><tr><td style=\"text-align:left\">demvoteshare_c</td><td>66.042<sup>***</sup></td><td>-5.683<sup>**</sup></td><td>1.403<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(3.160)</td><td>(2.609)</td><td>(0.044)</td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>30.508<sup>***</sup></td><td></td><td>0.526<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.817)</td><td></td><td>(0.010)</td></tr><tr><td style=\"text-align:left\">lagdemocrat:demvoteshare_c</td><td>-96.475<sup>***</sup></td><td></td><td>-0.849<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(3.852)</td><td></td><td>(0.049)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>13,577</td><td>13,577</td><td>13,577</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.267</td><td>0.434</td><td>0.749</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.267</td><td>0.434</td><td>0.749</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>27.944 (df=13573)</td><td>24.544 (df=13573)</td><td>0.246 (df=13573)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>1863.059<sup>***</sup> (df=3; 13573)</td><td>4160.728<sup>***</sup> (df=3; 13573)</td><td>25212.808<sup>***</sup> (df=3; 13573)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca62c259a0>"
+      ]
+     },
+     "execution_count": 17,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm_1 = lm_robust('score ~ lagdemocrat*demvoteshare_c', data = lmb_data)\n",
+    "lm_2 = lm_robust('score ~ democrat*demvoteshare_c', data = lmb_data)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat*demvoteshare_c', data = lmb_data)\n",
+    "print(\"Original results based on ADA Scores -- Full Sample with linear interactions\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "<ipython-input-18-668d3e6ccb56>:1: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n"
+     ]
+    }
+   ],
+   "source": [
+    "lmb_data['demvoteshare_sq'] = lmb_data['demvoteshare_c']**2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Original results based on ADA Scores -- Full Sample with linear and quadratic interactions\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>33.547<sup>***</sup></td><td>15.606<sup>***</sup></td><td>0.330<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.707)</td><td>(0.575)</td><td>(0.013)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>44.402<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(0.909)</td><td></td></tr><tr><td style=\"text-align:left\">democrat:demvoteshare_c</td><td></td><td>111.896<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(9.779)</td><td></td></tr><tr><td style=\"text-align:left\">democrat:demvoteshare_sq</td><td></td><td>-229.954<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(19.536)</td><td></td></tr><tr><td style=\"text-align:left\">demvoteshare_c</td><td>134.977<sup>***</sup></td><td>-23.850<sup>***</sup></td><td>2.798<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(9.634)</td><td>(6.711)</td><td>(0.193)</td></tr><tr><td style=\"text-align:left\">demvoteshare_sq</td><td>212.127<sup>***</sup></td><td>-41.729<sup>***</sup></td><td>4.294<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(22.395)</td><td>(14.672)</td><td>(0.447)</td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>13.030<sup>***</sup></td><td></td><td>0.322<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1.268)</td><td></td><td>(0.018)</td></tr><tr><td style=\"text-align:left\">lagdemocrat:demvoteshare_c</td><td>57.055<sup>***</sup></td><td></td><td>0.091<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(15.152)</td><td></td><td>(0.237)</td></tr><tr><td style=\"text-align:left\">lagdemocrat:demvoteshare_sq</td><td>-641.851<sup>***</sup></td><td></td><td>-8.804<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(30.802)</td><td></td><td>(0.508)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>13,577</td><td>13,577</td><td>13,577</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.371</td><td>0.456</td><td>0.822</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.370</td><td>0.456</td><td>0.822</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>25.892 (df=13571)</td><td>24.075 (df=13571)</td><td>0.207 (df=13571)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>1529.377<sup>***</sup> (df=5; 13571)</td><td>2589.018<sup>***</sup> (df=5; 13571)</td><td>89733.300<sup>***</sup> (df=5; 13571)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca62c3c8b0>"
+      ]
+     },
+     "execution_count": 19,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm_1 = lm_robust('score ~ lagdemocrat*demvoteshare_c + lagdemocrat*demvoteshare_sq', \n",
+    "                 data = lmb_data)\n",
+    "lm_2 = lm_robust('score ~ democrat*demvoteshare_c + democrat*demvoteshare_sq', \n",
+    "                 data = lmb_data)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat*demvoteshare_c + lagdemocrat*demvoteshare_sq', \n",
+    "                 data = lmb_data)\n",
+    "print(\"Original results based on ADA Scores -- Full Sample with linear and quadratic interactions\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "lmb_subset = lmb_data[lmb_data.demvoteshare.between(.45, .55)]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Results based on ADA Scores -- Close Sample with linear and quadratic interactions\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "<table style=\"text-align:center\"><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\"></td><tr><td style=\"text-align:left\"></td><td>(1)</td><td>(2)</td><td>(3)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align:left\">Intercept</td><td>37.121<sup>***</sup></td><td>21.437<sup>***</sup></td><td>0.418<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(0.969)</td><td>(1.817)</td><td>(0.013)</td></tr><tr><td style=\"text-align:left\">democrat</td><td></td><td>45.191<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(2.676)</td><td></td></tr><tr><td style=\"text-align:left\">democrat:demvoteshare_c</td><td></td><td>-688.343<sup>***</sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(247.490)</td><td></td></tr><tr><td style=\"text-align:left\">democrat:demvoteshare_sq</td><td></td><td>-3887.820<sup></sup></td><td></td></tr><tr><td style=\"text-align:left\"></td><td></td><td>(4798.097)</td><td></td></tr><tr><td style=\"text-align:left\">demvoteshare_c</td><td>830.925<sup>***</sup></td><td>450.846<sup>***</sup></td><td>15.699<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(20.936)</td><td>(161.201)</td><td>(0.227)</td></tr><tr><td style=\"text-align:left\">demvoteshare_sq</td><td>5333.335<sup>***</sup></td><td>7878.904<sup>***</sup></td><td>91.607<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(837.795)</td><td>(2992.602)</td><td>(10.882)</td></tr><tr><td style=\"text-align:left\">lagdemocrat</td><td>7.347<sup>***</sup></td><td></td><td>0.167<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1.588)</td><td></td><td>(0.020)</td></tr><tr><td style=\"text-align:left\">lagdemocrat:demvoteshare_c</td><td>-156.876<sup>***</sup></td><td></td><td>0.125<sup></sup></td></tr><tr><td style=\"text-align:left\"></td><td>(35.692)</td><td></td><td>(0.356)</td></tr><tr><td style=\"text-align:left\">lagdemocrat:demvoteshare_sq</td><td>-10116.678<sup>***</sup></td><td></td><td>-188.329<sup>***</sup></td></tr><tr><td style=\"text-align:left\"></td><td>(1433.721)</td><td></td><td>(16.325)</td></tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Observations</td><td>2,387</td><td>2,387</td><td>2,387</td></tr><tr><td style=\"text-align: left\">R<sup>2</sup></td><td>0.445</td><td>0.563</td><td>0.774</td></tr><tr><td style=\"text-align: left\">Adjusted R<sup>2</sup></td><td>0.444</td><td>0.562</td><td>0.774</td></tr><tr><td style=\"text-align: left\">Residual Std. Error</td><td>23.533 (df=2381)</td><td>20.885 (df=2381)</td><td>0.238 (df=2381)</td></tr><tr><td style=\"text-align: left\">F Statistic</td><td>469.571<sup>***</sup> (df=5; 2381)</td><td>617.595<sup>***</sup> (df=5; 2381)</td><td>6711.523<sup>***</sup> (df=5; 2381)</td></tr><tr><td colspan=\"4\" style=\"border-bottom: 1px solid black\"></td></tr><tr><td style=\"text-align: left\">Note:</td>\n",
+       " <td colspan=\"3\" style=\"text-align: right\">\n",
+       "  <sup>*</sup>p&lt;0.1;\n",
+       "  <sup>**</sup>p&lt;0.05;\n",
+       "  <sup>***</sup>p&lt;0.01\n",
+       " </td></tr></table>"
+      ],
+      "text/plain": [
+       "<stargazer.stargazer.Stargazer at 0x7fca442e1610>"
+      ]
+     },
+     "execution_count": 21,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "lm_1 = lm_robust('score ~ lagdemocrat*demvoteshare_c + lagdemocrat*demvoteshare_sq', \n",
+    "                 data = lmb_subset)\n",
+    "lm_2 = lm_robust('score ~ democrat*demvoteshare_c + democrat*demvoteshare_sq', \n",
+    "                 data = lmb_subset)\n",
+    "lm_3 = lm_robust('democrat ~ lagdemocrat*demvoteshare_c + lagdemocrat*demvoteshare_sq', \n",
+    "                 data = lmb_subset)\n",
+    "print(\"Results based on ADA Scores -- Close Sample with linear and quadratic interactions\")\n",
+    "Stargazer([lm_1, lm_2, lm_3])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "<ipython-input-22-75fbc0c4726e>:4: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "<ipython-input-22-75fbc0c4726e>:7: SettingWithCopyWarning: \n",
+      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
+      "Try using .loc[row_indexer,col_indexer] = value instead\n",
+      "\n",
+      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/layer.py:467: PlotnineWarning: geom_point : Removed 39 rows containing missing values.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781669006282)>"
+      ]
+     },
+     "execution_count": 22,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#aggregating the data\n",
+    "lmb_data = lmb_data[lmb_data.demvoteshare.between(.45, .55)]\n",
+    "categories = lmb_data.lagdemvoteshare\n",
+    "lmb_data['lagdemvoteshare_100'] = pd.cut(lmb_data.lagdemvoteshare, 100)\n",
+    "\n",
+    "agg_lmb_data = lmb_data.groupby('lagdemvoteshare_100')['score'].mean().reset_index()\n",
+    "lmb_data['gg_group'] = [1 if x>.5 else 0 for x in lmb_data.lagdemvoteshare]\n",
+    "agg_lmb_data['lagdemvoteshare'] = np.arange(0.01, 1.01, .01)\n",
+    "#plotting\n",
+    "\n",
+    "p.ggplot(lmb_data, p.aes('lagdemvoteshare', 'score')) +\\\n",
+    "    p.geom_point(p.aes(x = 'lagdemvoteshare', y = 'score'), data = agg_lmb_data) +\\\n",
+    "    p.stat_smooth(p.aes('lagdemvoteshare', 'score', group = 'gg_group'), \n",
+    "                  data=lmb_data, method = \"lm\", \n",
+    "              formula = 'y ~ x + I(x**2)') +\\\n",
+    "    p.xlim(0,1) + p.ylim(0,100) +\\\n",
+    "    p.geom_vline(xintercept = 0.5)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/stats/smoothers.py:310: PlotnineWarning: Confidence intervals are not yet implementedfor lowess smoothings.\n",
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/layer.py:467: PlotnineWarning: geom_point : Removed 39 rows containing missing values.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781668991251)>"
+      ]
+     },
+     "execution_count": 23,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(lmb_data, p.aes('lagdemvoteshare', 'score')) +\\\n",
+    "    p.geom_point(p.aes(x = 'lagdemvoteshare', y = 'score'), data = agg_lmb_data) +\\\n",
+    "    p.stat_smooth(p.aes('lagdemvoteshare', 'score', group = 'gg_group'), \n",
+    "                  data=lmb_data, method = \"lowess\") +\\\n",
+    "    p.xlim(0,1) + p.ylim(0,100) +\\\n",
+    "    p.geom_vline(xintercept = 0.5)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/tcaputo/opt/anaconda3/lib/python3.8/site-packages/plotnine/layer.py:467: PlotnineWarning: geom_point : Removed 39 rows containing missing values.\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 640x480 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "<ggplot: (8781675201482)>"
+      ]
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "p.ggplot(lmb_data, p.aes('lagdemvoteshare', 'score')) +\\\n",
+    "    p.geom_point(p.aes(x = 'lagdemvoteshare', y = 'score'), data = agg_lmb_data) +\\\n",
+    "    p.stat_smooth(p.aes('lagdemvoteshare', 'score', group = 'gg_group'), \n",
+    "                  data=lmb_data, method = \"lm\")+\\\n",
+    "    p.xlim(0,1) + p.ylim(0,100) +\\\n",
+    "    p.geom_vline(xintercept = 0.5)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Smoothing and Density [WIP]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Questions\n",
+    "- Can you think of another example where you might use the close election design to estimate some average treatment effect?\n",
+    "- To what degree does this study help us understand the importance of incumbency in a Presidential election?  Why/why not?"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  },
+  "toc": {
+   "base_numbering": 1,
+   "nav_menu": {},
+   "number_sections": true,
+   "sideBar": true,
+   "skip_h1_title": false,
+   "title_cell": "Table of Contents",
+   "title_sidebar": "Contents",
+   "toc_cell": false,
+   "toc_position": {},
+   "toc_section_display": true,
+   "toc_window_display": false
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/Python/Synthetic_Control.ipynb b/Python/Synthetic_Control.ipynb
new file mode 100644
index 0000000..b81ed54
--- /dev/null
+++ b/Python/Synthetic_Control.ipynb
@@ -0,0 +1,356 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Welcome\n",
+    "\n",
+    "This is material for the **Synthetic Control** chapter in Scott Cunningham's book, [Causal Inference: The Mixtape.](https://mixtape.scunning.com/)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
+    "\n",
+    "from rpy2 import robjects\n",
+    "from rpy2.robjects import pandas2ri\n",
+    "pandas2ri.activate()\n",
+    "from rpy2.robjects.vectors import IntVector\n",
+    "\n",
+    "import plotnine as p\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "%matplotlib inline\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# read data\n",
+    "def read_data(file):\n",
+    "    return pd.read_stata(\"https://raw.github.com/scunning1975/mixtape/master/\" + file)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Prison Construction and Black Male Incarceration\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "texas = read_data(\"texas.dta\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from rpy2.robjects.packages import importr\n",
+    "from rpy2.robjects.conversion import localconverter\n",
+    "Synth = importr('Synth')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "X1, X0, Z1, Z0 all come directly from dataprep object.\n",
+      "\n",
+      "\n",
+      "**************** \n",
+      " searching for synthetic control unit  \n",
+      " \n",
+      "\n",
+      "**************** \n",
+      "**************** \n",
+      "**************** \n",
+      "\n",
+      "MSPE (LOSS V): 2019839 \n",
+      "\n",
+      "solution.v:\n",
+      " 0.06843543 0.02849309 0.8949852 3.30278e-05 0.000122374 0.007929645 1.2876e-06 \n",
+      "\n",
+      "solution.w:\n",
+      " 2.1036e-06 1.638e-07 1.2605e-06 8.781e-06 0.3420653 3.717e-07 1.469e-07 1.663e-07 3.97e-08 0.2963684 2.978e-07 1.866e-07 6.0836e-06 6.135e-07 1.2125e-06 9.987e-07 5.646e-07 5.5707e-06 0.3613969 1.0905e-06 2.03e-07 2.069e-07 3.0885e-06 3.587e-07 4.5e-09 9.953e-07 4.9594e-06 6.338e-07 2.669e-07 2.468e-07 2.597e-07 4.057e-06 3.443e-07 7.378e-07 3.7061e-06 6.8669e-06 2.7571e-06 7.46e-07 1.136e-06 3.282e-07 2.3e-09 2.8444e-06 8.1658e-05 7.937e-07 3.366e-07 3.824e-07 2.04033e-05 7.517e-07 7.518e-07 \n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "control_units = [1, 2, 4, 5, 6] +\\\n",
+    "    list(range(8, 14)) + list(range(15,43)) +\\\n",
+    "    list(range(44, 47)) + [49, 50, 51, 53,54,55,56]\n",
+    "\n",
+    "robjects.globalenv['texas'] = texas\n",
+    "\n",
+    "predictors = robjects.vectors.StrVector(['poverty', 'income'])\n",
+    "sp = robjects.vectors.ListVector({'1': ['bmprison', IntVector([1988, 1990, 1991, 1992]), 'mean'], \n",
+    "                                  '2': ['alcohol', 1990, 'mean'], \n",
+    "                                  '3': ['aidscapita', IntVector([1990, 1991]), 'mean'], \n",
+    "                                  '4': ['black', IntVector([1990, 1991, 1992]), 'mean'], \n",
+    "                                  '5': ['perc1519', 1990, 'mean']})\n",
+    "\n",
+    "dataprep_out = Synth.dataprep(texas, \n",
+    "    predictors = predictors,\n",
+    "    predictors_op=\"mean\",\n",
+    "    time_predictors_prior=np.arange(1985, 1994),\n",
+    "    special_predictors=sp,\n",
+    "    dependent='bmprison',\n",
+    "    unit_variable='statefip',\n",
+    "    unit_names_variable='state',\n",
+    "    time_variable='year',\n",
+    "    treatment_identifier=48,\n",
+    "    controls_identifier=control_units,\n",
+    "    time_optimize_ssr=np.arange(1985, 1994),\n",
+    "    time_plot=np.arange(1985, 2001))\n",
+    "\n",
+    "synth_out = Synth.synth(data_prep_obj = dataprep_out)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>ct_weights</th>\n",
+       "      <th>statefip</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>2.103617e-06</td>\n",
+       "      <td>1</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>1.638053e-07</td>\n",
+       "      <td>2</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>1.260496e-06</td>\n",
+       "      <td>4</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>8.781002e-06</td>\n",
+       "      <td>5</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>3.420653e-01</td>\n",
+       "      <td>6</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "     ct_weights  statefip\n",
+       "0  2.103617e-06         1\n",
+       "1  1.638053e-07         2\n",
+       "2  1.260496e-06         4\n",
+       "3  8.781002e-06         5\n",
+       "4  3.420653e-01         6"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "weights = synth_out.rx['solution.w'][0]\n",
+    "ct_weights = pd.DataFrame({'ct_weights':weights.flatten(), 'statefip':control_units})\n",
+    "ct_weights.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "texas = pd.merge(ct_weights, texas, how='right', on='statefip')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Synthetic Control Performance')"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "texas = texas.sort_values('year')\n",
+    "ct = texas.groupby('year').apply(lambda x : np.sum(x['ct_weights']*x['bmprison']))\n",
+    "treated = texas[texas.statefip==48]['bmprison'].values\n",
+    "years = texas.year.unique()\n",
+    "\n",
+    "\n",
+    "plt.plot(years, ct, linestyle='--', color='black', label='control')\n",
+    "plt.plot(years, treated, linestyle='-', color='black', label='treated')\n",
+    "plt.ylabel('bmprison')\n",
+    "plt.xlabel('Time')\n",
+    "plt.title('Synthetic Control Performance')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Treated - Control')"
+      ]
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "ct_diff = treated - ct\n",
+    "\n",
+    "plt.plot(years, np.zeros(len(years)), linestyle='--', color='black', label='control')\n",
+    "plt.plot(years, ct_diff, linestyle='-', color='black', label='treated')\n",
+    "plt.ylabel('bmprison')\n",
+    "plt.xlabel('Time')\n",
+    "plt.title('Treated - Control')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "\n",
+    "#### Questions\n",
+    "- In your own words, what do you think the identifying assumptions are for synthetic control to be consistent? \n",
+    "- What role, if any, does parallel trends play in synthetic control?\n",
+    "- Who is the unit with the largest ratio of post to pre RMSPE?  \n",
+    "- Compare the unit with the largest post to pre RMSPE estimated effect to the Texas effect.  How do the weights compare?  How do the size of the effects compare?  How do the ``signs`` of the effects compare?\n",
+    "- Can you improve on my fit by experimenting with different combinations? Do so and report your analysis.\n",
+    "- Report results from a variety of different specifications.  How robust does the prison effect appear to be?\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}