Abstract #300157

Back to main JSM 2001 Program page

Add To My Program  /  View My Program  /  What is My Program?

JSM 2001 Abstract #300157
Activity Number: 259
Type: Invited
Date/Time: Wednesday, August 8, 2001 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Sports*
Abstract - #300157
Title: An Intelligent Bayesian Approach to College Football Rankings
Author(s): Ed Kambour*+
Affiliation(s): PROS
Address: 3223 Smith Street, Houston, Texas, 77006,

One of the hot topics in sports, particularly College football, is the use of "computer" ratings systems. The advent of the Bowl Championship Series has heightened interest. A least squares estimator of ratings will be introduced. This method will be extended to a generalized least squares (GLS) method that accounts for heterogeniety and time-series effects. The GLS method will enhanced by including prior information involving the teams' past performance. The resulting Bayesian approach utilizes Markov Chain Monte Carlo Methods, specifically Gibbs Sampling. Results from past NCAA Division I seasons will be used as examples.

  • The address information is for the authors that have a + after their name.
  • Authors who are presenting talks have a * after their name.

Back to the full JSM 2001 program

JSM 2001

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2001