JSM 2004 - Toronto

Abstract #301254

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); 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 2004 Program page
Activity Number: 437
Type: Contributed
Date/Time: Thursday, August 12, 2004 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #301254
Title: The Use of James-Stein Estimators in Tests of Homogeneity of the Risk Difference
Author(s): Purnima Rao-Melacini*+ and Colleen Kelly
Companies: McMaster University and San Diego State University
Address: 584 Governer's Road, Dundas , ON, L9H 5E3, Canada
Keywords: multicenter data ; treatment difference ; meta-analysis ; institution effect ; shrinkage estimators
Abstract:

The evaluation of the effectiveness of a new treatment as compared to previous therapies requires the design of a clinical trial. In this context, a multicenter clinical experiment is one in which two or more institutions agree to follow the same protocol. In medical and pharmaceutical research, responses are often measured on a dichotomous scale and then efficacy may be measured with the risk difference. To summarize the risk difference across centers, the estimated risk difference for each center must be comparable. Lipsitz et al. (1998) have proposed several tests of homogeneity of the risk difference in sets of 2x2 tables for sparse data, to which Lui and Kelly (2000) made several important changes and recommendations. We investigate using James-Stein estimates in weighted least squares statistics of homogeneity. These estimates shrink the individual proportions towards the overall mean and thus avoid the problems encountered when risk estimates are zero or one. In Monte Carlo simulations, we see that these estimates have a lower mean squared error with respect to the MLE, particularly when the mean sample size per treatment is small.

  • 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 2004 program

JSM 2004 For information, contact jsm@amstat.org or phone (888) 231-3473. If you have questions about the Continuing Education program, please contact the Education Department.
Revised March 2004