Abstract #300521

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 #300521
Activity Number: 97
Type: Contributed
Date/Time: Monday, August 4, 2003 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract - #300521
Title: Analysis of Incomplete Longitudinal Binary Data: Comparing Multiple Imputation with GEE and a Naïve Approach
Author(s): Xiaoming Li*+ and Devan V. Mehrotra and John Barnard
Companies: Merck & Co., Inc. and Merck Research Laboratories and DeCODE Genetics
Address: UNA-102, Blue Bell, PA, 19422,
Keywords: longitudinal binary data ; missing data ; multiple imputation ; GEE
Abstract:

Consider a randomized trial in which subjects receive either vaccine A or B. Blood is drawn at fixed time points and assayed for immune activity. Each assay result is classified as being either positive ("responder") or negative. We need to estimate the true difference (A-B) in the proportion of responders at time T, and test the hypothesis that it is zero. However, due to incomplete follow-up, data at time T are not available for all subjects. In statistical terms, we have a problem of incomplete longitudinal binary data. An easy solution is to use only the available data at time T ("naïve" method). An alternative is to use the generalized estimating equations (GEE) approach for longitudinal data. We propose a multiple imputation (MI) method to tackle the incomplete data problem, and demonstrate three key results via simulation. First, if data are missing completely at random, MI is substantially more efficient than the naïve method, and slightly more efficient than GEE. Second, with small samples, GEE often fails due to "convergence problems," but MI is free of that problem. Finally, if the data are missing at random, unlike the naïve method and GEE, MI yields unbiased results.

  • 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