|
Activity Number:
|
385
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 11, 2004 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #300889 |
|
Title:
|
Modeling Multivariate Longitudinal Profiles
|
|
Author(s):
|
Geert Verbeke*+ and Steffen Fieuws
|
|
Companies:
|
Katholieke Universiteit Leuven and Biostatistical Centre, K.U.Leuven
|
|
Address:
|
Biostatistical Centre, U.Z. Sint Rafael, Leuven, B-3000, Belgium
|
|
Keywords:
|
random effects ; multivariate ; linear mixed models ; association
|
|
Abstract:
|
Multivariate longitudinal data arise when a set of different outcomes on the same unit is measured repeatedly over time. There are different situations where a joint modeling approach is needed. First, the association structure can be of importance. A possible question might be how the association between outcomes evolves over time or if outcome-specific evolutions are related to each other. In a second situation, the aim is to improve the results of a discriminant analysis by using more than one longitudinally measured outcome. In another situation, interest lies on estimation of the fixed effects. As an example, consider testing the difference in evolution between many outcomes. Linear mixed models are a flexible tool for joint modeling purposes, especially when the gathered data are unbalanced. However, computational problems due to the dimension of the joint covariance matrix of the random effects arise as soon a the number of outcomes and/or the number of used random effects increases. A pseudo-likelihood approach will be presented to circumvent this problem. Real data will be used to illustrate the approach.
|
- 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 2004 program |
