Workshop 1 – Vasishth / Nicenboim

Introduction to Bayesian Modeling using Stan

Shravan Vasishth (with Bruno Nicenboim), University of Potsdam

In this one-day workshop, we will give a comprehensive introduction to using Stan for Bayesian data analysis and Bayesian modeling.

By the end of the course, participants should be able to:

  1. Understand the syntax of Stan and the high-level ideas behind MCMC and HMC.
  2. Fit standard models (such as linear models) in Stan.
  3. Understand how hierarchical modeling works, and be able to fit complex hierarchical models in different settings.
  4. Carry out sensitivity analyses to investigate how posteriors change as a result of prior specification.
  5. Visualize and interpret different models.
  6. Carry out posterior predictive checks and cross-validation for model evaluation.

We will provide lecture notes and suggested readings for further study. We assume that everyone has a laptop with them and has the R package rstan installed within R.

This one-day workshop will involve lectures interspersed with short exercises to be done in class. In order to consolidate understanding, we will assign a project that participants can carry out (this is optional). Students have the option to submit it to the instructor a week later and get feedback.

More information on the workshop are provided here: http://www.ling.uni-potsdam.de/~vasishth/courses/IntroStanFGME2017.html