Just Another Gibbs Sampler - PhD Project


The main goal of this document is to introduce the usage of JAGS as a software to perform Bayesian statistics without having to implement many lines of codes. JAGS offers an interface script where all the specification related to the data, model, priors and hyper- parameters are carried out. In addition, in order to exemplify its applicability, we provide a set of examples involving linear regression, generalized linear models, generalized linear mixed models and survival models. All the examples are implemented both in R and Python through interfaces provided by the packages R2jags and pyjags, respectively.