Abstract #300705

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 #300705
Activity Number: 460
Type: Invited
Date/Time: Thursday, August 7, 2003 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #300705
Title: A Noniterative Robust Approach for Analyzing Repeated Measures Designs
Author(s): Devan V. Mehrotra*+ and Xiaoyin Frank Fan
Companies: Merck Research Laboratories and Merck & Co., Inc.
Address: Clinical Biostatistics, UN-A102, Blue Bell, PA, 19422,
Keywords: longitudinal study ; correlated data ; missing data ; variance components ; outliers
Abstract:

Iterative methods such as restricted maximum likelihood (REML), the default in SAS PROC MIXED, are now routinely used to analyze repeated measures designs. REML, which is equivalent to ANOVA for a balanced design, has attractive large sample properties for normal data but performs poorly under non-normality. We present a noniterative robust alternative to REML for repeated measures analysis. Specifically, we develop formulas for robust estimates of variance components and fixed effect parameters (cell means), and use them to test common hypotheses of interest. A key element of our proposal is to replace standard (nonrobust) means and variances with one-step M-estimates of location and A-estimates of scale, respectively. Simulations reveal that our proposed method, called RAVE, is almost as efficient as REML for normal data, but significantly outperforms REML when sampling from either a heavy-tailed distribution or from a mixture of normal distributions. A numerical example is presented for illustration.

  • 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