Abstract #300999

This is the preliminary program for the 2003 Joint Statistical Meetings in San Francisco, California. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 2-5, 2003); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2003 Program page
JSM 2003 Abstract #300999
Activity Number: 371
Type: Contributed
Date/Time: Wednesday, August 6, 2003 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract - #300999
Title: Evaluating Predictive Accuracy of CART and MARS Models for Gene-Gene Interactions
Author(s): Nancy R. Cook*+
Companies: Brigham and Women's Hospital
Address: 900 Commonwealth Ave. East, Boston, MA, 02215-1204,
Keywords: prediction ; exploratory data analysis ; data mining ; cross-validation
Abstract:

Causes of complex diseases may be multifactorial, and interactions among genes, or epistasis, may play an important biologic role. Models such as classification and regression trees (CART) and multivariate adaptive splines (MARS) can assist in detecting and examining potential genetic interactions. Both methods offer advantages over typical logistic regression methods in that they may discover interactions of genes that do not display strong main effects. Because of the great potential for overfitting with these as well as other exploratory methods, however, assessment of model fit cannot be based on training data nor on test data used for model selection, both of which may seriously overestimate model fit. Use of external data or cross-validation including the entire selection process is required to provide unbiased estimates. We use these methods to explore the presence of gene-gene interactions in ischemic stroke in data from a 92 SNP panel of candidate genes from the Physicians' Health Study. We use cross-validation to evaluate and compare the performance of the methods in predicting ischemic stroke in these data.

  • 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 2003 program

JSM 2003 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 2003