|
Activity Number:
|
310
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
WNAR
|
| Abstract - #307741 |
|
Title:
|
Optimal Designs for Linear Mixed Effects Models
|
|
Author(s):
|
Martijn P.F. Berger*+
|
|
Companies:
|
University of Maastricht
|
|
Address:
|
P.O. Box 616, Maastricht, International, 6200 MD, Netherlands
|
|
Keywords:
|
longitudinal designs ; linear mixed effects models ; optimal design ; D-optimality ; maximin
|
|
Abstract:
|
Longitudinal designs have been used to study effects over time in various different fields. These designs are characterized by the number and allocation of time points at which measurements are taken and by the number of distinct cohorts of people that are sampled. The most suitable models to analyze these data are the linear mixed effects models, because they enable random variation among and within subjects and cohorts. Efficient estimation of the parameters in these models depends on the characteristics of the longitudinal design. In this presentation we will discuss some of the problems that are involved in finding optimal designs for these models. D-optimal designs for these longitudinal designs will be presented and we will show how much efficiency is lost if sub-optimal designs are used. Guidelines for efficient designing of longitudinal studies will also be offered.
|
- 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 |
