Introduction to Bayesian Statistics (357-0-20)
Instructors
Martin A Tanner
847/491-2700
2006 Sheridan Road
Meeting Info
STAT Sem Rm B02 - 2006 Sher: Wed, Fri 9:30AM - 10:50AM
Overview of class
The purpose of this course is to provide an elementary introduction to a variety of computational algorithms for the Bayesian analysis of data. Two types of methods are considered in detail: observed data and data augmentation methods. The observed data methods are applied directly to the likelihood or to the posterior distribution. These include: Newton-Raphson, Monte Carlo and Metropolis methods. The data augmentation methods rely on an augmentation of the data which simplifies the likelihood or posterior distribution. These include: EM, Data Augmentation, and the Gibbs sampler. All methods are motivated and illustrated with real examples. Overall, this course provides the student with a good introduction to the field, the ability to read application papers, and the ability to apply these methods to problems of interest to the student. Students understand at a heuristic level how the methods work and when a given method may be preferred over another.
Registration Requirements
Stat 320-1,2,3 and Stat 350 are prerequisites
Teaching Method
Lecture
Evaluation Method
Homework assignments and final exam
Class Materials (Required)
1. Probability and Bayesian Modeling by Albert and Hu.
https://www.routledge.com/Probability-and-Bayesian-Modeling/Albert-Hu/p/book/9781138492561#
2. Tools for Statistical Inference: Methods for the Exploration of Posterior Distributions and Likelihood Functions by M. Tanner THIRD EDITION ONLY required http://www.springer.com/us/book/9780387946887
3. Bayesian Methods by Jeff Gill THIRD EDITION ONLY required http://www.crcpress.com/Bayesian-Methods-A-Social-and-Behavioral-Sciences-Approach-Third-Edition/Gill/p/book/9781439862483
Class Materials (Suggested)
Bayesian Computation with R by Jim Albert SECOND EDITION ONLY required http://www.springer.com/us/book/9780387922973
Class Notes
This course is computationally intensive and a working knowledge of programs such as R or Python is expected.
Class Attributes
Formal Studies Distro Area