Introduction to Bayesian Modeling

Bayesian methods are reshaping how scientists reason about uncertainty, with applications spanning biostatistics, social/behavioral science, deep learning, and many more. This workshop offers an in-depth, hands-on introduction to the Bayesian modeling. Prior knowledge of R will be assumed and RStan will be introduced in this workshop. Over three focused sessions, we will discuss the comparison of Bayesian and frequentist paradigms through engaging real-world examples, then build up the core machinery you need to model your own data. Topics include the choice of priors, posterior inference, a brief introduction to MCMC, model checking, linear regression & generalized linear models in a Bayesian framework (logistic regression), and hierarchical models (these kinds of models are where Bayesian methods and MCMC shine). Every concept is paired with practical implementation in R and RStan. We'll end this workshop with a discussion of the pitfalls of current MCMC methods and a brief introduction at variational inference as a rising, scalable alternative and its applications.