This package is built to implement the Gibbs-Metropolis-Hastings Hybrid Algorithm for Bayesian meta-analysis of multi-trial odds ratios with a normally distributed random effect.
-The hierarchical analysis uses modularized functions, data pregeneration, and matrix opterations to increase efficiency.
-The bayesian analysis uses rcpp and rcpp armadillo to increase efficiency.
This package was modeled off of the template by Patrick Muchmore.
You can install the development version of gmhs2 from GitHub by cloning the repository and building the package.
This R package includes use for empirical and hierarchical bayesian meta analysis of the logit model with normal random effects. Please see the Bayesian Meta Analysis Vignette for a tutorial.