Introduction to Bayesian Inference: Core Principles and Application in Stan
(PhD students can get 2 ECTS)
The Department of Psychology.
Dr. Daniel Heck, University of Mannheim, Germany.
Date: Wednesday 8.05.2019 meeting-room no. 03.4.409, 4th floor, and Thursday 9.05.2019, canteen 03.2.M202, 2nd floor, floor), Department of Psychology, Øster Farimagsgade 2A, DK-1353 Copenhagen
8th and 9th May 2019, 9:00 – 17:00, lunch 12:00 – 13:00.
Wednesday 8.05.2019 meeting-room no. 03.4.409, 4th floor.
Thursday 9.05.2019, each day 10-13h and 14-17th,in the canteen 03.2.M202, 2nd floor.
Department of Psychology, Øster Farimagsgade 2A, DK-1353 Copenhagen.
Please click here to register for the course
Deadline for registration: 2nd May 2019.
Minimum 10, maximum 20 participants.
This 2-day course introduces the core principles of Bayesian inference and illustrates their application using the software Stan. The first day will focus on the logic of Bayesian parameter estimation and the conceptual foundations of Markov chain Monte Carlo methods. Practical examples show how to fit simple models in Stan. The second day introduces the Bayes factor for testing hypotheses and its theoretical and pragmatic benefits. Examples show how to compute and interpret the Bayes factor, for instance, for a t-test with default priors.
PhD students and members of the Department of Psychology can participate in this seminar. The entire seminar is held in English. A maximum of 20 participants will be allowed in this course. Basic statistical knowledge is required (e.g., about the standard t-test and the binomial distribution), but you do not need to have any prior knowledge on Bayesian statistics. Moreover, basic skills in R are required (e.g., working with vectors, calling functions, etc.). If you are not familiar with R, you may learn the basics prior to the course. A good introduction is available in “Chapter 2: An introduction to R” in the freely available book: https://learningstatisticswithr.com/lsr-0.6.pdf
Participants should bring their notebook with a recent installation of the R environment (available from www.r-project.org) and RStudio (available from www.rstudio.com). Moreover, it is necessary to install the R packages “rstan” and “BayesFactor”. The former requires the R development tools (for instructions how to install these, see https://support.rstudio.com/hc/en-us/articles/200486498-Package-Development-Prerequisites ). If you have questions about any of the prerequisites, please do not hesitate to contact the instructor via email. Course materials (e.g., slides) will be provided at the beginning of the course.
There are many good articles on Bayesian inference. As a start, I recommend the following introductions:
- Bayesian inference as optimal knowledge-updating process:
Wagenmakers, E.-J., Morey, R. D., & Lee, M. D. (2016). Bayesian benefits for the pragmatic researcher. Current Directions in Psychological Science, 25(3), 169–176. doi:10.1177/0963721416643289
- Theoretical and pragmatic benefits:
Wagenmakers, E.-J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., Love, J., … Morey, R. D. (2018). Bayesian inference for psychology. Part I: Theoretical advantages and practical ramifications. Psychonomic Bulletin & Review, 25(1), 35–57. doi:10.3758/s13423-017-1343-3
- Introduction to formal details of Bayes’ rule:
Etz, A., & Vandekerckhove, J. (2018). Introduction to Bayesian inference for psychology. Psychonomic Bulletin & Review, 25(1), 5–34. doi:10.3758/s13423-017-1262-3