36-735 Methods for Correlated Data and Bayesian Statistics 1

6 units, Carnegie Mellon University

This course will provide a broad overview of methods for the analysis of correlated data. Topics include generalized linear models, hierarchical models, and generalized estimating equations. We will cover the fundamentals of Bayesian statistics including posterior and predictive distributions, conjugate priors, maximum a posteriori estimators, Laplace approximations, and Markov Chain Monte Carlo algorithms. Students will conduct data analyses using the probabilistic programming language Stan as well as R packages such as lme4. On a subset of topics students will derive and implement methods from scratch. A strong understanding of probability theory, multivariable calculus, and linear algebra is required.

Prerequisites: none listed

Courseweave needs JavaScript to show sections, workloads and planning tools.