ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

ECPR

Install the app

Install this application on your home screen for quick and easy access when you’re on the go.

Just tap Share then “Add to Home Screen”

Basket

You don't have anything in your basket.

Bayesian Modelling

Chendi Wang
chendi.wang@eui.eu

European University Institute

Chendi is a research fellow at the European University Institute (EUI), working in the SOLID ERC project. He also holds a PhD in political science from the EUI.

Chendi's research interests lie in political behaviour, political economy, public policy and quantitative and computational methods.

He has published in the British Journal of Political Science, Western European Politics, and in volumes published by Cambridge University Press.


Course Dates and Times

Monday 6 – Friday 10 February 2023
2.5 hours of live teaching per day
09:30 - 12:00noon CET

Prerequisite Knowledge

This course requires basic knowledge in statistical analysis, including linear regression models and hypothesis testing. Some exposure to models with limited dependent variables (e.g. binary) is also required.

If you do not have this knowledge, take Introduction to Inferential Statistics or Applied Regression Analysis.

JAGS (or BUGS) and Stan will be used through R, and therefore we would prefer you to have basic knowledge of R, though it is not absolutely necessary. If you are completely new to R, consider taking Introduction to R.

Knowledge of maximum likelihood estimation (MLE) is an asset but not a prerequisite.


Short Outline

This course provides a highly interactive online teaching and learning environment, using state-of-the-art online pedagogical tools. It is designed for a demanding audience (researchers, professional analysts, advanced students) and capped at a maximum of 16 participants so that the teaching team can cater to the specific needs of each individual.

Purpose of the course

The course will teach you to understand and apply various Bayesian methods for answering research questions in quantitative social science. In addition to the theoretical material, you will gain proficiency in data analytic skills by using the open-source statistical programming language R.

This is an advanced course for students who already have basic quantitative methods training.

By the end of the course you will:

  • understand the fundamental differences and similarities between frequentist and Bayesian approaches to inference
  • be able to formulate linear and generalised linear models, hierarchical models and measurement models in the Bayesian framework using JAGS or Stan
  • know how to interpret models in the Bayesian framework.

You will also be able to compare and assess Bayesian models and apply the Bayesian methods to political science research questions.

The course is suitable for researchers, professional analysts, and advanced students.

ECTS Credits

3 credits Engage fully with class activities 
4 credits Complete a post-class assignment


Long Course Outline

To get the most out of this course, complete the required in-depth readings for each day, and skim at least one of the recommended readings, if listed.

Five pre-recorded lectures, introducing the course's major topics and concepts, supplement the readings.

Key topics covered

Monday

Why Bayesian? Bayesian inference concepts, simulation-based inference and MCMC.

Tuesday

The linear model, and models for binary outcome.

Wednesday

Discrete choice outcomes and count outcomes.

Thursday

Hierarchical models and measurement models.

Friday

Model assessment and comparison.

How the course will work online

The course combines pre-class readings and pre-recorded videos with daily two-hour live Zoom sessions. These sessions focus on two tasks:

  • in-depth discussion of topics covered in the lecture
  • a lab session with guided hands-on exercises relevant to the lecture.

The lab sessions will enable you to master the technical side of Bayesian modelling in R. These sessions will also enable you to apply the statistical methods we discuss during lectures to real-world data.

The Instructor will distribute the R script in advance so you can explore the code at your own pace, but we will go through the code and models together during the sessions.

We will get to know each other, and each other's projects, and explore how we can apply Bayesian modelling to answer your research questions. There will also be problem sets after each session. We will discuss these assignments, and any problems you may have, together the following day.

You can share thoughts and ask questions on our Slack channel, and the Instructor will host live Q&A sessions and social breaks. You will be able to sign up for a quick one-to-one consultation during designated office hours.


Additional Information

Disclaimer

This course description may be subject to subsequent adaptations (e.g. taking into account new developments in the field, participant demands, group size, etc). Registered participants will be informed at the time of change.

By registering for this course, you confirm that you possess the knowledge required to follow it. The instructor will not teach these prerequisite items. If in doubt, please contact us before registering.