Abstract #302197

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 #302197
Activity Number: 299
Type: Contributed
Date/Time: Tuesday, August 5, 2003 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract - #302197
Title: Validation and Calibration of Proportional Hazards Models
Author(s): Somesh Chattopadhyay*+ and Daniel L. McGee
Companies: Duke University and Florida State University
Address: 2711 Allen Rd., Apt. D6, Tallahassee, FL, 32312,
Keywords: calibration ; validation ; prognostic model ; proportional hazards model
Abstract:

Whether a prognostic model based on one population is applicable to other populations is of interest to both epidemiologists and clinicians. We present an empirical example of model validation and calibration based on person-level data from 22 studies that include both sexes, black and white groups, and samples from the U.S. and other countries. The database analyzed includes almost 250,000 individuals with almost 14,000 deaths from coronary heart disease occurring during follow-up. We develop a proportinal hazards model based on one study and examine whether it is applicable to the other 21 studies. We demonstrate several aspects of the model validation process including calibrating the model through a linear transformation of the model at the appropriate scale. Among the other possibilities, we examine whether combining models from multiple studies, accompanied by calibration performs better than depending on the model from a single study. We will examine these possibilities in context of survival models for death from coronary heart disease.

  • 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