This course is designed to provide an introduction to fundamental conceptual, computational, and practical methods of Bayesian data analysis. The first part will introduce the Bayesian approach, including

  • Comparison with frequentist methods
  • Bayesian learning
  • Common prior distributions
  • Summarizing posterior distributions

In order to study problems with more than a few parameters, modern Bayesian computing algorithms are required. In the second part of the course we will study these methods, mostly MCMC and mostly using R. The main topics include

  • Monte Carlo approximation
  • Gibbs sampling
  • Convergence diagnostics
  • JAGS

With these computational tools at hand, we will begin applying Bayesian methods using

  • Multiple linear regression
  • Generalized linear models
  • Hierarchical models

We will pay special attention to comparing models with each other and testing for model adequacy.

For more information, view the course syllabus.