Advanced Bayesian Statistics Using R (edX)

This course is directed at people who are already familiar with the fundamentals of Bayesian inference. It explores further the concepts, methods, and algorithms introduced in the part one (Introductory Bayesian Data Analysis Using R).
The course places mixed effects regression models useful for experiments with repeated measures or additional hierarchy often encountered in biostatistics, ecology and health sciences among others within the Bayesian context. It takes a closer look at the Markov Chain Monte Carlo (MCMC) algorithms, why they work and how to implement them in the R programming language. Convergence assessment and visualisation of the results are discussed in some detail. The course also explores Bayesian model averaging, often used in machine learning, all within the context of practical examples.
Finally, we discuss different kinds of missing data, and the Bayesian methods of dealing with such situations.
Prior facility in basic algebra and calculus as well as programming in R is highly recommended.
This course is part of the Bayesian Statistics Using R Professional Certificate.

What you'll learn
• Using latent (unobserved) variables and dealing with missing data.
• Multivariate analysis within the context of mixed effects linear regression models. Structure, assumptions, diagnostics and interpretation. Posterior inference and model selection.
• Why Monte Carlo integration works and how to implement your own MCMC Metropolis-Hastings algorithm in R.
• Bayesian model averaging in the context of change-point problem. Pinpointing the time of change and obtaining uncertainty