Event location:
Aula IV, Department of Statistical Sciences, Via Belle Arti 41 (STAT PhD Classes Virtual Room on need)
Type:
Cycle 41 - Elective Courses
Aim: The course introduces the theoretical foundations of Bayesian modeling, starting from the Bayesian linear model and extending to hierarchical models and Generalised Linear Models. A practical component focuses on the use of the probabilistic programming language Stan to fit complex and customizable models via Hamiltonian Monte Carlo. The course also presents Bayesian tools for model checking, model comparison, and variable selection, providing practical guidance for applied statistical modelling.
Learning outcome: At the end of the course, students will be able to plan and implement complex Bayesian models and estimate them using modern computational tools. They will know how to interpret results from posterior distributions, assess model assumptions through appropriate Bayesian model-checking techniques, and identify possible extensions of basic modelling frameworks to address practical research problems.
Course contents
- The Bayesian linear model and Bayesian GLMs
- Introduction to hierarchical models
- Introduction to the Stan probabilistic programming language
- The posterior predictive distribution and its role in model checking