MATH 375 - Probability & Bayesian Statistics II

3 credits
Bayesian statistical modeling and inference for data analysis is examined with an emphasis on model likelihoods and prior and posterior distributions. Common models for both univariate and multivariate data based on binomial, beta, gamma, exponential and normal distributions will be examined. Topics include model fit assessment and comparison; single- and multi-parameter models; and Bayesian computation methods: Markov-chain Monte-Carlo simulation, Gibbs sampler, and the Metropolis-Hastings algorithm. Additional topics may include theory of conjugate priors and missing data imputation techniques. This course is the second of a two-course sequence and requires coding using appropriate statistical software.

Prerequisite(s): MATH 262 and MATH 275 with a grade of C- or higher; and CMSC 140 or CMSC 160
Corequisite(s): None