Prerequisites: STAT 308 or STAT 408.
A graduate-level treatment of Bayesian methods for data analysis focusing on the theoretical foundations and computational strategies for modern Bayesian inference. Topics include the formulation and analysis of single- and multi-parameter Bayesian models, hierarchical and multilevel structures, and Bayesian generalized linear models. The course will also consider advanced Markov Chain Monte Carlo computational techniques such as Gibbs sampling, Metropolis-Hastings algorithm, and Hamiltonian Monte Carlo. Emphasis is placed on both the mathematical underpinnings of Bayesian analysis and the implementation of complex models using contemporary software tools.
Outcomes: By the end of the course, students will be able to: formulate and analyze complex Bayesian models; compare and select complex Bayesian models; critically interpret and communicate Bayesian analyses in research settings; extend basic Bayesian methods to advanced/nonstandard models.
A graduate-level treatment of Bayesian methods for data analysis focusing on the theoretical foundations and computational strategies for modern Bayesian inference. Topics include the formulation and analysis of single- and multi-parameter Bayesian models, hierarchical and multilevel structures, and Bayesian generalized linear models. The course will also consider advanced Markov Chain Monte Carlo computational techniques such as Gibbs sampling, Metropolis-Hastings algorithm, and Hamiltonian Monte Carlo. Emphasis is placed on both the mathematical underpinnings of Bayesian analysis and the implementation of complex models using contemporary software tools.
Outcomes: By the end of the course, students will be able to: formulate and analyze complex Bayesian models; compare and select complex Bayesian models; critically interpret and communicate Bayesian analyses in research settings; extend basic Bayesian methods to advanced/nonstandard models.