Logo Utrecht University

Multilevel Conference

Key note 3: Paul Bürkner

Title: Bayesian Multilevel Modeling with brms and Stan

Abstract:
The talk will be about Bayesian multilevel models and their implementation in R using the brms package. I will provide a brief introduction to Bayesian statistics in general as well as to Stan, which is a highly flexible probabilistic programming language to fit open-ended Bayesian models. I will then explain how to access Stan using R formula syntax via brms. It supports a wide range of response distributions and modeling options such as splines, autocorrelation, or censoring all in a multilevel context. Several post-processing methods are implemented to visualize results, predict new data, or compare and select models. Some examples from Psychology and Medicine will be discussed.