|
Activity Number:
|
222
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #303619 |
|
Title:
|
Cost-Efficient Designs for Longitudinal Mixed Effects Models
|
|
Author(s):
|
Martijn Berger*+
|
|
Companies:
|
Maastricht University
|
|
Address:
|
P.O. Box 616, Maastricht, International, 6200 MD, The Netherlands
|
|
Keywords:
|
Optimal Design ; Mixed Effects ; Longitudinal ; Costs ; Random Parameters ; Efficiency
|
|
Abstract:
|
Linear mixed effects models are often applied to analyze time-structured data. Optimal designs for estimating fixed parameters in these models depend on the covariance structure of the errors and the random parameters. Since the costs of sampling subjects may differ from the costs of obtaining repeated measurements, an optimal design may not be the most cost-efficient one. In this presentation I will compare different designs in terms of their costs. I use a cost function that distinguishes the costs of sampling subjects from the costs of obtaining repeated measurements and compare the cost efficiency of different designs for different mixed effects models. An example is used to show that it is sometimes beneficial to increase the number of distinct time points instead of increasing the sample size when the relative costs of sampling subjects becomes high.
|
- 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 2009 program |
