Statistics: Bayesian Statistics

Introduction to Bayesian inference. Posterior inference, predictive distributions, prior distribution selection. MCMC algorithms. Hierarchical models. Model checking and selection. Use of computer software. Prerequisite: Content knowledge of STAT 5510 assumed.

Content knowledge of STAT 5510 assumed; Open to Degree and PACE students

Bayesian analysis are now widely used in academia and industry. This course is an introduction to state-of-the-art Bayesian data analysis. Topics include: Review of conditional probabilities, posterior inference, predictive distributions, prior distribution selection. MCMC algorithms. Hierarchical models. Model checking and selection. Use of computer software for Bayesian statistics.

Pre/co-requisite: Content knowledge of STAT 5510 assumed. Familiarity with at least one programming language (e.g. R, python, Julia, Matlab) is highly recommended. Learning objectives: 1. Bayesian data analysis By the end of this class, you will be able to prepare and present state-of-the-art Bayesian analysis of data sets relevant to your interests or research. 2. Scientific Skills. A significant component of the class will be the final project. In doing this project, you will improve your ability to, learn on your own, communicate complex statistical ideas to a diverse audience, and write state-of-the-art statistical algorithms.

Problem sets (40%), project (50%) and quizzes (10%). No exams.