From 3a81a0a24a151fc1eba4154edebac681e438ca43 Mon Sep 17 00:00:00 2001 From: lrjohnson0 Date: Mon, 22 Jul 2024 21:01:28 -0400 Subject: [PATCH] finalized intro lecture --- VB_IntroTimeDepData.qmd | 60 +++++++++++++++++++++++++---------- docs/VB_IntroTimeDepData.html | 59 ++++++++++++++++++++++------------ docs/search.json | 10 +++--- materials_temp.qmd | 6 ++-- 4 files changed, 90 insertions(+), 45 deletions(-) diff --git a/VB_IntroTimeDepData.qmd b/VB_IntroTimeDepData.qmd index 5303264..1deb320 100644 --- a/VB_IntroTimeDepData.qmd +++ b/VB_IntroTimeDepData.qmd @@ -50,21 +50,37 @@ insert_html_math <- make_latex_decorator("", "$$") ## classoption: aspectratio=169 ``` -## Learning Objectives - -1. Understand basic concepts in how we understand and model time-dependent population data in VBD applications -2. Review basic idea of forecasting -3. Overview of course ## Population dynamics of disease -The number of hosts, vectors, pathogens, and infected individuals change over time +The number of hosts, vectors, pathogens, and infected individuals change over time. -We use models to understand and to forecast/predict +We use models models of various types to: +- understand relationships +- forecast/predict possible future outcomes ## What types of models? -tactical to strategic +
+ +![](graphics/model_continuum.png) + +
+ +::: columns +::: {.column width="50%"} + +- Focus on describing/quantifying patterns +- Statistical Models, e.g. regression, time series + +::: + +::: {.column width="50%"} +- Seek to understand mechanisms, less prediction +- ODEs, dynamical systems, Stochastic DEs, Individual Based Models + +::: +::: We'll focus on the tactical end of things here (i.e., no dif eqs) @@ -153,7 +169,9 @@ for(j in 1:length(sigma)){ ## Observation Models -We also have to go out into the field and take some observations of the populations. Let's say that we observe $N_{\mathrm{obs}}(t)$ individuals at time $t$. How does this relate to the true population size? One possibility is: \begin{align*} +We also have to go out into the field and take some observations of the populations. + +Let's say that we observe $N_{\mathrm{obs}}(t)$ individuals at time $t$. How does this relate to the true population size? One possibility is: \begin{align*} N_{\mathrm{obs}}(t) = N(t) + V(t) \end{align*} where $V(t)$ is our "observation uncertainty", and all together this equation describes our `r myblue("observation model")`. @@ -162,19 +180,29 @@ N_{\mathrm{obs}}(t) = N(t) + V(t) With the observation model, our full system (process + observation models) might be \begin{align*} N(t) & = s N(t-1) + b(t-1) +W(t)\\ N_{\mathrm{obs}}(t) & = N(t) + V(t) -\end{align*} plus distributions for $W(t)$, $V(t)$, and the initial population size. +\end{align*} + +plus: + +- distributions for $W(t)$, $V(t)$ +- the initial population size. ## Observation process matters -Analysis approach depends on the sampling -- evenly or unevenly spaced, goal of the modeling exercise (understanding vs forecasting/prediction). +Analysis approach depends on not only the goal of the modeling exercise (understanding vs forecasting/prediction) but also on details of the way that data are observed + +- evenly or unevenly spaced observations +- consistent sampling locations +- types of observation/instrument (automated? trap type?) -## Forecasting process +Although we focus on tools this week, we'll also talk about general modeling considerations and how to think some of these as we conduct an analyses. -something about forecasting ## Outline of Course -1. Abundance data from VecDyn and NEON -2. Regression refresher for time dep data -- basics plus time dependent predictors, transformations, simple AR +0. Introductions and Goals +1. Regression refresher focusing on diagnostics and transformations +2. Regression approaches for time dependent data -- basics plus time dependent predictors, transformations, simple AR 3. Analysis of evenly-spaced data: basic time-series methods -4. Advanced modeling with Gaussian Process Models +4. Abundance data from VecDyn and NEON + climate and meteorological variables +5. Advanced modeling with Gaussian Process Models diff --git a/docs/VB_IntroTimeDepData.html b/docs/VB_IntroTimeDepData.html index 3c89d9d..3833ff1 100644 --- a/docs/VB_IntroTimeDepData.html +++ b/docs/VB_IntroTimeDepData.html @@ -337,23 +337,30 @@

VectorByte Methods Training

- -
-

Learning Objectives

-
    -
  1. Understand basic concepts in how we understand and model time-dependent population data in VBD applications
    2. -
  2. Review basic idea of forecasting
    3. -
  3. Overview of course
    4. -

Population dynamics of disease

-

The number of hosts, vectors, pathogens, and infected individuals change over time

-

We use models to understand and to forecast/predict

+

The number of hosts, vectors, pathogens, and infected individuals change over time.

+

We use models models of various types to: - understand relationships - forecast/predict possible future outcomes

What types of models?

-

tactical to strategic

+
+

+
+
+
+
    +
  • Focus on describing/quantifying patterns
    • +
  • Statistical Models, e.g. regression, time series
    • +
+
+
    +
  • Seek to understand mechanisms, less prediction
    • +
  • ODEs, dynamical systems, Stochastic DEs, Individual Based Models
    • +
+
+

We’ll focus on the tactical end of things here (i.e., no dif eqs)

@@ -399,7 +406,8 @@

A first (stochastic) model

Observation Models

-

We also have to go out into the field and take some observations of the populations. Let’s say that we observe \(N_{\mathrm{obs}}(t)\) individuals at time \(t\). How does this relate to the true population size? One possibility is: \[\begin{align*} +

We also have to go out into the field and take some observations of the populations.

+

Let’s say that we observe \(N_{\mathrm{obs}}(t)\) individuals at time \(t\). How does this relate to the true population size? One possibility is: \[\begin{align*} N_{\mathrm{obs}}(t) = N(t) + V(t) \end{align*}\] where \(V(t)\) is our “observation uncertainty”, and all together this equation describes our observation model.

@@ -408,22 +416,31 @@

Observation Models

With the observation model, our full system (process + observation models) might be \[\begin{align*} N(t) & = s N(t-1) + b(t-1) +W(t)\\ N_{\mathrm{obs}}(t) & = N(t) + V(t) -\end{align*}\] plus distributions for \(W(t)\), \(V(t)\), and the initial population size.

+\end{align*}\]

+

plus:

+

Observation process matters

-

Analysis approach depends on the sampling – evenly or unevenly spaced, goal of the modeling exercise (understanding vs forecasting/prediction).

-
-
-

Forecasting process

-

something about forecasting

+

Analysis approach depends on not only the goal of the modeling exercise (understanding vs forecasting/prediction) but also on details of the way that data are observed

+ +

Although we focus on tools this week, we’ll also talk about general modeling considerations and how to think some of these as we conduct an analyses.

Outline of Course

-
    -
  1. Abundance data from VecDyn and NEON
    2. -
  2. Regression refresher for time dep data – basics plus time dependent predictors, transformations, simple AR
    3. +
      +
    1. Introductions and Goals
      2. +
    2. Regression refresher focusing on diagnostics and transformations
      3. +
    3. Regression approaches for time dependent data – basics plus time dependent predictors, transformations, simple AR
    4. Analysis of evenly-spaced data: basic time-series methods
      5. +
    5. Abundance data from VecDyn and NEON + climate and meteorological variables
    6. Advanced modeling with Gaussian Process Models
    diff --git a/docs/search.json b/docs/search.json index ee28402..55a1531 100644 --- a/docs/search.json +++ b/docs/search.json @@ -11,14 +11,14 @@ "href": "VB_IntroTimeDepData.html#population-dynamics-of-disease", "title": "VectorByte Methods Training", "section": "Population dynamics of disease", - "text": "Population dynamics of disease\nThe number of hosts, vectors, pathogens, and infected individuals change over time\nWe use models to understand and to forecast/predict" + "text": "Population dynamics of disease\nThe number of hosts, vectors, pathogens, and infected individuals change over time.\nWe use models models of various types to: - understand relationships - forecast/predict possible future outcomes" }, { "objectID": "VB_IntroTimeDepData.html#what-types-of-models", "href": "VB_IntroTimeDepData.html#what-types-of-models", "title": "VectorByte Methods Training", "section": "What types of models?", - "text": "What types of models?\ntactical to strategic\nWe’ll focus on the tactical end of things here (i.e., no dif eqs)" + "text": "What types of models?\n\n\n\n\n\n\nFocus on describing/quantifying patterns\nStatistical Models, e.g. regression, time series\n\n\n\nSeek to understand mechanisms, less prediction\nODEs, dynamical systems, Stochastic DEs, Individual Based Models\n\n\n\nWe’ll focus on the tactical end of things here (i.e., no dif eqs)" }, { "objectID": "VB_IntroTimeDepData.html#simple-example-a-first-deterministic-model", @@ -39,14 +39,14 @@ "href": "VB_IntroTimeDepData.html#observation-models", "title": "VectorByte Methods Training", "section": "Observation Models", - "text": "Observation Models\nWe also have to go out into the field and take some observations of the populations. Let’s say that we observe \\(N_{\\mathrm{obs}}(t)\\) individuals at time \\(t\\). How does this relate to the true population size? One possibility is: \\[\\begin{align*}\nN_{\\mathrm{obs}}(t) = N(t) + V(t)\n\\end{align*}\\] where \\(V(t)\\) is our “observation uncertainty”, and all together this equation describes our observation model." + "text": "Observation Models\nWe also have to go out into the field and take some observations of the populations.\nLet’s say that we observe \\(N_{\\mathrm{obs}}(t)\\) individuals at time \\(t\\). How does this relate to the true population size? One possibility is: \\[\\begin{align*}\nN_{\\mathrm{obs}}(t) = N(t) + V(t)\n\\end{align*}\\] where \\(V(t)\\) is our “observation uncertainty”, and all together this equation describes our observation model." }, { "objectID": "VB_IntroTimeDepData.html#observation-process-matters", "href": "VB_IntroTimeDepData.html#observation-process-matters", "title": "VectorByte Methods Training", "section": "Observation process matters", - "text": "Observation process matters\nAnalysis approach depends on the sampling – evenly or unevenly spaced, goal of the modeling exercise (understanding vs forecasting/prediction)." + "text": "Observation process matters\nAnalysis approach depends on not only the goal of the modeling exercise (understanding vs forecasting/prediction) but also on details of the way that data are observed\n\nevenly or unevenly spaced observations\nconsistent sampling locations\ntypes of observation/instrument (automated? trap type?)\n\nAlthough we focus on tools this week, we’ll also talk about general modeling considerations and how to think some of these as we conduct an analyses." }, { "objectID": "VB_IntroTimeDepData.html#forecasting-process", @@ -60,7 +60,7 @@ "href": "VB_IntroTimeDepData.html#outline-of-course", "title": "VectorByte Methods Training", "section": "Outline of Course", - "text": "Outline of Course\n\nAbundance data from VecDyn and NEON\nRegression refresher for time dep data – basics plus time dependent predictors, transformations, simple AR\nAnalysis of evenly-spaced data: basic time-series methods\nAdvanced modeling with Gaussian Process Models" + "text": "Outline of Course\n\nIntroductions and Goals\nRegression refresher focusing on diagnostics and transformations\nRegression approaches for time dependent data – basics plus time dependent predictors, transformations, simple AR\nAnalysis of evenly-spaced data: basic time-series methods\nAbundance data from VecDyn and NEON + climate and meteorological variables\nAdvanced modeling with Gaussian Process Models" }, { "objectID": "GP_Solutions.html", diff --git a/materials_temp.qmd b/materials_temp.qmd index 22bfaed..d0100a7 100644 --- a/materials_temp.qmd +++ b/materials_temp.qmd @@ -38,9 +38,9 @@ We are assuming familiarity with R basics as well as at least introductory stati # Live Workshop Materials -## Introduction to the Workshop and Time Dependent Abundance Data +## Introduction to the Workshop -- [Lecture (coming soon)]() +- [Lecture](VB_IntroTimeDepData.qmd)

    @@ -76,7 +76,7 @@ We are assuming familiarity with R basics as well as at least introductory stati - [The VecDyn website](https://vectorbyte.crc.nd.edu/vecdyn-datasets). - [About the VecDyn API](https://www.vectorbyte.org/blog/vecdyn-api) -- Other Materials Coming Soon. +- Other Materials Coming Soon!