|
Activity Number:
|
143
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 4, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #301777 |
|
Title:
|
Comparison of Designs for Correlated Response Models with an Unknown Dispersion Matrix
|
|
Author(s):
|
Sourish Saha*+ and Andre Khuri
|
|
Companies:
|
GlaxoSmithKline and University of Florida
|
|
Address:
|
, Collegeville, PA, ,
|
|
Keywords:
|
correlated response models ; dispersion matrix ; power of multivariate tests ; quantile dispersion graphs
|
|
Abstract:
|
This talk deals with the comparison of designs for correlated response models with an unknown variance-covariance matrix. The novelty of the proposed approach lies in combining the "Quantile Dispersion Graphs (QDGs)" technique with the power of the multivariate test for the equality of parameters from such models. The approach is based on considering the quantiles of a certain criterion function (namely, power of three multivariate tests) on concentric surfaces within a particular region of the so-called alternative space. The dependence of these quantiles on the unknown values of the variances and covariances obtained from the variance-covariance matrix, is depicted by plotting the so-called QDGs of the criterion function. These plots provide a clear assessment of the magnitude of the power value associated with a given design. A numerical example is presented for illustration.
|
- 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 2008 program |
