Abstract #300679

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 #300679
Activity Number: 249
Type: Contributed
Date/Time: Tuesday, August 5, 2003 : 10:30 AM to 12:20 PM
Sponsor: Biometrics Section
Abstract - #300679
Title: Combining Stratified and Unstratified Logrank Tests for Matched Pairs Survival Data
Author(s): David Oakes*+ and Changyong Feng
Companies: University of Rochester and University of Kansas Medical Center
Address: Dept. of Biostatistics - Box 630, Rochester, NY, 14642-0001,
Keywords: proportional hazards ; censored data ; Mantel test ; Hougaard model ; frailty ; life-tables
Abstract:

We consider testing for a treatment effect in matched pairs survival data, with one member of each pair receiving active treatment, the other receiving the control. The stratified logrank test counts the "preferences" established by each pair. The unstratified logrank test is optimal under the proportional hazards model under within-pair independence, but not under dependence. Moreover the variance estimate for the logrank statistic requires adjustment due to the within-pair dependence. While the stratified logrank test and unstratified logrank tests are optimal (under proportional hazards models) for extreme dependence and independence respectively, in intermediate cases, a linear combination of the two statistics may be locally more powerful than either individual statistic. Under Hougaard's positive stable frailty model, we derive the optimal linear combination and show how to estimate it from the data. We show that for moderate dependence this combined test statistic is more powerful than either individual statistic. We examine the robustness of the procedure to the choice of frailty distribution and briefly consider extensions to blocks of arbitrary size.

  • 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