Please note that this workshop will be held on 16th - 17th April and 23rd-24th April! Please go to the schedule for more details.  

Bayesian methods have become increasingly central to the social sciences and many adjacent fields. Once a niche methodology with steep computational and software entry barriers, Bayesian statistics is now a readily accessible toolbox available through user-friendly interfaces in R. Yet, despite the growing popularity of these methods, many researchers remain unsure when and why to use Bayesian approaches, how to interpret their results, and how to move beyond standard pre-implemented model types toward custom solutions.

 

This four-day workshop, spread across two weeks, provides an integrated path from foundations to custom applications of Bayesian modelling. The first half introduces participants to the core ideas and workflow of Bayesian statistics, juxtaposing Bayesian and frequentist perspectives, covering estimation, inference, and prediction, and demonstrating how to use the R package brms for applied Bayesian modelling with minimal coding effort. The second half then opens the black box: participants learn how to build their own Bayesian models in Stan, the probabilistic programming language underlying brms. Through a combination of conceptual input, guided lab sessions, and hands-on coding exercises, participants will gain both the theoretical grounding and practical skills to understand, use, and design Bayesian models for their own research questions - ranging from straightforward applications to fully custom model specifications.

 

By the end of the course, participants will be familiar with the Bayesian workflow, fluent in interpreting and diagnosing Bayesian models in R, and capable of implementing tailored modelling solutions in Stan that go beyond the limits of off-the-shelf packages.

 

Organizational structure of the course

This course is organised in four days, each consisting of three sessions à 90 minutes. A 90-minute lunch break divides morning lecture sessions and afternoon lab sessions.

Lectures will alternate between classical lecturing, interactive Q&A sessions, and regular applied code demonstrations and empirical illustrations.

Labs allow you to work collaboratively on applied problem sets in small groups in Zoom breakout rooms. The problem sets typically comprise a sequence of problems as typically encountered in real-world empirical work. Hints will be provided, and the instructor will be available in the main meeting room to assist with any questions that may arise. Toward the end of the session, we will jointly discuss the solutions to the problem sets.

Software requirements

This workshop requires installations of recent versions of R, RStudio, and RStan. Throughout the second half of the workshop, we will use Stan via its R interface RStan. Setting up RStan can be somewhat time-consuming as it requires the installation of a C++ compiler. Workshop participants should follow these instructions on the Stan Development Team's GitHub to install and configure the rstan package and its prerequisites on their operating system before the workshop. If you do not have administrator privileges on the machine you intend to use, please approach your system administrator in advance of the workshop.