|
Activity Number:
|
529
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
| Abstract - #304067 |
|
Title:
|
Invariance in Multivariate Linear Models with High Dimension, Low Sample Size
|
|
Author(s):
|
Yueh-Yun Chi*+ and Keith E. Muller
|
|
Companies:
|
University of Florida and University of Florida
|
|
Address:
|
1329 SW 16th Street, Gainesville, FL, 32610,
|
|
Keywords:
|
MANOVA ; Canonical correlation ; Discriminant ; Orthogonal ; Affine ; Orthonormal
|
|
Abstract:
|
High throughput technologies in medical imaging, genetics, and chemical analysis generate an ever increasing number of outcome variables for each independent sampling unit. The singularity of residual covariance matrices precludes using most multivariate analysis methods and destroys affine invariance. Generalized inverses provide a straightforward solution, and advantageously, also provides many of the invariance properties in full rank problems. We describe the invariance properties in multivariate linear models that do and do not change when the number of variables exceeds the sample size. The properties hold important promise in the development of multivariate theory for high dimension and low sample size data.
|
