SO-IAA School on Bayesian Statistics

During five half days the school will cover topics from the basics of Bayesian inference over model fitting, computational methods to simulations.

Practical exercises will be combined with the theoretical lectures

The agenda will include:

• Basics of Bayesian Inference. A quick review of Bayes theorem, the basics of probability and probability distributions, followed by a fun exercise that introduces Bayesian inference in practice and how to define prior distributions. Finally, some connection to maximum likelihood and linear regression.
• Bayesian Computation. Bayesian inference in practice: the basics of sampling algorithms, and how to know whether your model is less wrong than other models.
• Beyond the linear model: hierarchical and generalized linear models. Going beyond the basics: population inference with hierarchical Bayesian modelling, fast inference with probabilistic and differentiable programming, and going beyond linear regression
• Simulation-based Bayesian Inference. Bayesian inference with implicit likelihoods: how to sample posteriors when the likelihood is not accessible, but a simulator is.

Professors: Gwen Eadie (University of Toronto) and Daniela Huppenkothen (SRON, Netherlands)