STAT 615 Advanced Bayesian Methods
This course meets
- TR 11-12:20 @ East 0111
Relevant course pages
Complex hierarchical and multilevel models, dynamic linear and generalized linear models, spatial models. Bayesian nonparametric methods. Specialized Markov chain Monte Carlo algorithms and practical approaches to increasing mixing and speed convergence. Summarizing posterior distributions, and issues in inference. Model assessment, model selection, and model averaging.
Student Learning Outcomes
- Understand the common Bayesian methods used for time series analysis, spatial analysis, shrinkage, and nonparametrics.
- Read, comprehend, and implement methods in the primary literature.
- Apply these methods within your own research agenda.
Please use the Blackboard discussion forum.
The course aims to provide an overview of Bayesian methods in a variety of areas including some of the following:
- Hierarchical (linear) models
- Ridge regression
- Hamiltonian Monte Carlo
- Dynamic models
- Dynamic linear models
- State-space models
- Sequential Monte Carlo
- Gaussian process models
- Conditional autogressive models
- Slice sampling
- Finite mixture models
- Dirichlet process
- Advanced computing (?)
- Adaptive rejection Metropolis sampling
- Integrated nested Laplace approximation
Faculty Senate Recommendations
This course abides by the Faculty Senate Recommendations provided at http://www.celt.iastate.edu/teaching/preparing-to-teach/recommended-iowa-state-university-syllabus-statements.