Courses Detail Information
STAT4510J – Bayesian Analysis
Instructors:
Credits:
4 credits
Pre-requisites:
ECE4010J
Description:
This course provides an introduction to Bayesian statistical modeling and inference, emphasizing both the theoretical foundations and computational methods of modern Bayesian data analysis. Students will learn how to construct probabilistic models, update beliefs with data, and perform inference using posterior distributions.
The course covers key topics such as prior specification, posterior computation, Bayes factors, hierarchical models, and decision-theoretic principles, alongside computational algorithms including Markov Chain Monte Carlo (MCMC) and variational inference. Through hands-on work in R and Python, students will develop the skills to implement Bayesian methods and apply them to complex real-world data problems.
Course Topics:
1. Review of statistical inference and probability
2. Bayesian foundations and simple Bayesian models
3. Prior and posterior distributions; conjugacy and model updating
4. Posterior simulation and analysis, Bayesian computation including MCMC methods
5. Hierarchical Bayesian models and multilevel structures
6. Bayesian approaches to missing data problems
7. Model comparison, Bayes factors, and predictive checks
8. Modern applications in data science, machine learning, and applied research