EECS 800 Statistics Learning

Instructor	Text Book	Grades
Course Description	Exam	Lectures
Prerequisites	Project	Useful Links

Instructor

Dr. Xue-wen Chen

Ø Phone: (785)864-8825

Ø Email: xwchen AT ku DOT edu

Ø Office: Eaton 2028

Ø Lectures: MWF, 9:30 am – 10:20 am

Ø Location: Room 2111 Learned Hall

Ø Class Number: 14291

Ø Office Hours: MW: 10:40 am – 11:40 am or by appointment

Back to top

Course Description

This course introduces Bayesian inference, Bayesian modeling and computation in statistics. A variety of topics such as hierarchical models, Markov chain simulation, Gibbs sampler, and MCMC will be covered. Applications in bioinformatics will also be discussed.

Back to top

Prerequisites

Ø Mathematical maturity, e.g., knowing to use algebra in solving problems.

Ø Knowledge in basic probability and statistics.

Ø Some experience with computer programming (MATLAB or C or C++).

Back to top

Text Book

Ø Andrew Gelman, John B. Carlin, Hal S. Stern, and Donald B. Rubin. Bayesian Data Analysis. 2^nd Ed., 2004. ISBN 1-58488-388-X, Chapman & Hall/CRC.

Ø Additional handouts will be provided in class as needed.

Back to top

Exam

Ø One closed-book exam will be conducted in class.

Back to top

Project

Students will form groups to complete a course project. Each group consists of 2 ~ 3 students and chooses a topic that pertains to the course material. The topics can be theoretical derivations or analyses, applications of statistical learning methods to problems you are interested in, or something else. Be creative.

Each group should be formed and submit a pre-proposal by October 31. The pre-proposal should describe the project goal, the problems to be studied, current methods, proposed methods, expected results, references, names of the group members and responsibilities of each member (about 5 pages: single space, fond size = 12, references are not counted).

In addition to the issues addressed in pre-proposal, the final project should include your results and discussions. A written final report in the style of a journal article is also required. Final project is due by Dec. 12. Both pre-proposal and the final project should be typed (either with LATEX or with word processing). Late written project will not be accepted.

Each group will give classroom presentation about the final project.

Back to top

Grades

Course grade will be assigned based on scores on midterm exam, pre-proposal, final project, and presentation. The final grade is made up of the following:

Ø Midterm exam: 30%

Ø Pre-proposal: 15%

Ø Final Project: 40%

Ø Presentation 15%

The cutoffs for grades will be roughly as follows:

A: 90 – 100 B: 80 – 89 C: 70 – 79 D: 60 – 69 F: 0 – 59

Back to top

Lectures

 
Part I: Review/Introduction: Basic probability and Statistics
Ø One random variable 
Ø Multiple random variable
Ø Statistical inference (p-value, hypothesis test etc.)
Ø Stochastic processes etc.
 
Part II: Bayesian Data Analysis
Ø Single parameter models 
Ø Multiparameter models 
Ø Hierarchical models
Ø Computation issues 
Ø Markov chain simulation 
Ø Gibbs sampler 
Ø MCMC 
Ø Regression models
Ø Mixture models, etc.
 
Part III: Bioinformatics Applications
I will select some bioinformatics papers that employed the course material for class discussions.
Back to top

Useful Links

Ø Reading List on Bayesian Methods

Ø Introduction to MCMC

Ø The BUGS Project (Bayesian Inference Using Gibbs Sampling)

Ø Software for Flexible Bayesian Modeling and Markov Chain Sampling

Ø Bayesian Output Analysis Program

Ø The R Project for Statistical Computing

Back to top