|
Activity Number:
|
182
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
SSC
|
| Abstract - #308119 |
|
Title:
|
Reducing the Bias of Between- Within-Cluster Covariate Methods When Data Are Missing at Random
|
|
Author(s):
|
John Neuhaus*+ and Charles E. McCulloch
|
|
Companies:
|
University of California, San Francisco and University of California, San Francisco
|
|
Address:
|
185 Berry Street, Lobby 4, Suite 5700, San Francisco, CA, 94107-1762,
|
|
Keywords:
|
Covariate decompositions ; generalized linear mixed models ; conditional likelihood
|
|
Abstract:
|
Generalized linear mixed models that partition covariates into between-and within-cluster components can provide effective analysis of longitudinal data in settings where covariates or responses are missing completely at random. However, like conditional likelihood methods, such between/within cluster approaches can yield inconsistent covariate effect estimates when data are missing at random. This talk describes and evaluates several strategies, including weighted methods, to reduce bias when data are missing at random. We illustrate these methods with simulation studies and fits to example data.
|
- 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 2007 program |
