Program

7th LCSR Summer School «Bayesian Approach in Social Science» 

Instructor: Hannes Kröger, Socio-Economic Panel Study (SOEP), German Institute for Economic Research (DIW), Berlin, Germany; European University Institute, Florence, Italy

Day 1 — Introduction

09:30-10:00 / Registration
10:00-11:30 / Introduction; Philosophical differences between Bayes and Freq.
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / The conditional probability of an event: Should Peter worry about his cancer diagnosis?
13:20-14:30 / Lunch
14:30-15:30 / Some unique features of Bayesian statistics: Prior and Posterior distribution
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / How do prior, likelihood and posterior relate to each other?

Day 2 —  Bayesian regression and the use of prior information

10:00-11:30 / Examples of equivalency between Bayes and Frequentist, Introduction to Stan
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / More convenient ways to use Stan
13:20-14:30 / Lunch 
14:30-15:30 / Using Bayesian statistics to account for measurement error
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Beyond simple versions of measurement error, two variables measured with error

Day 3 — Application 1: Including prior knowledge – Small samples and identification problems

10:00-11:30 / Limited data as a structural problem, what does Bayes offer as solution?
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Including prior information from a previous study using small samples
13:20-14:30 / Lunch 
14:30-15:30 / Updating current knowledge with results from a new study
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Updating current knowledge with results from a new study 

Day 4 — A brief look on Bayesian Estimation

10:00-11:30 / Closed form solutions and the advantage of MCMC
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Constructing an own tiny MCMC
13:20-14:30 / Lunch
14:30-15:30 / Diagnostics of model convergence
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / A deterministic alternative to MCMC: INLA
17:00-18:30 / Consultations for Summer School participants (optional)

Day 5 — Application 2: Escaping the significance trap: Using degrees of support instead of tests and Bayesian updating of evidence

10:00-11:30 / The misuse of statistical significance testing
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Second generation p-values, type M and S errors
13:20-14:30 / Lunch
14:30-15:30 / Answering more interesting research questions in the Bayesian framework I
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Answering more interesting research questions in the Bayesian framework II
17:00-18:30 / Consultations for Summer School participants (optional)

Day 6 — Bayesian Evaluation of Informative Hypotheses (BEIH)

10:00-11:30 / What are informative hypotheses, how do we use them?
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / BEIH I: Do dissociative identity disorder patients have amnesia?
13:20-14:30 / Lunch
14:30-15:30 / Second example of BEIH
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Summing up the virtues of the posterior distribution
16:45-17:00 / Coffee break (room 212)
17:00-18:30 / Special Lecture by Ronald Inglehart «The Silent Revolution in Reverse: The Rise of Trump and the Xenophobic Populist Parties»

Day 7 — Model checking and selection

10:00-11:30 / How do I know if my model is good model? Which of two models is the better one?
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Example of fitting competing models
13:20-14:30 / Lunch 
14:30-15:30 / Identifying problems in models I
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Identifying problems in models II
17:00-18:30 / Consultations for Summer School participants (optional)

Day 8 — Bayesian Model Averaging and Multiple Bias modelling

10:00-11:30 / Different sources of bias, how do we integrate them?
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Bayesian multiple bias modelling
13:20-14:30 / Lunch 
14:30-15:30 / Different model specifications, model averaging versus scenario plotting I
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Different model specifications, model averaging versus scenario plotting II
17:00-18:30 / Consultations for Summer School participants (optional)
18:30-21:00 / Reception

Day 9 — Bayesian approach to measurement invariance and multilevel modelling

10:00-11:30 / Approximate Bayesian measurement invariance as an alternative to exact MI
11:30-11:50 / Coffee break (room 212)
11:50-13:20 / Measurement invariance of human values across countries in ESS
13:20-14:30 / Lunch
14:30-15:30 / Bayesian Multilevel models I
15:30-15:45 / Coffee break (room 212)
15:45-16:45 / Bayesian Multilevel models II
16:45-17:00 / Coffee break (room 212)
17:00-20:00 / Participants' presentations

Day 10 — Final Session

09:00-11:00 / Summary of the Summer School