BAYESIAN STATISTICS
Introduces the theory, methods, and computational procedures needed to perform advanced Bayesian data analyses. Predictive and decision-theoretic motivations including subjective probability, risk, admissibility, and exchangeability; highlights key components of Bayesian analysis (i.e., prior, likelihood, posterior, and predictive distributions) within standard parametric models and advanced hierarchical and multilevel models; demonstrates the iterative process of model specification, implementation, criticism, and revision with applied case studies; implements computational techniques (e.g., Markov chain Monte Carlo, variational inference) in modern probabilistic programming languages.
Prerequisites & course information
Requisites: STAT 609 , 610 , 611 , STAT/MATH 709 , ECON 709 , POLI SCI 818 , or COMP SCI/E C E 761 Course Designation: Grad 50% - Counts toward 50% graduate coursework requirement Repeatable for Credit: No Last Taught: Spring 2026 Learning Outcomes: 1. Justify the use of probability for coherent uncertainty quantification Audience: Graduate 2. Explain how Bayesian updating occurs in conjugate models and hierarchical models Audience: Graduate 3. Compare and contrast the conceptual and practical benefits and challenges of different posterior approximation strategies like MCMC and variational inference Audience: Graduate 4. Implement posterior approximation algorithms in modern statistical and probabilistic programming languages such as R or Stan Audience: Graduate 5. Specify, fit, criticize, and revise Bayesian models in practice Audience: Graduate Learning Outcomes: 1. Justify the use of probability for coherent uncertainty quantification Audience: Graduate 2. Explain how Bayesian updating occurs in conjugate models and hierarchical models Audience: Graduate 3. Compare and contrast the conceptual and practical benefits and challenges of different posterior approximation strategies like MCMC and variational inference Audience: Graduate 4. Implement posterior approximation algorithms in modern statistical and probabilistic programming languages such as R or Stan Audience: Graduate 5. Specify, fit, criticize, and revise Bayesian models in practice Audience: Graduate
Sources
Catalogue variants
- BAYESIAN STATISTICS · 3 credits.