Activity Number:
|
363
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Biometrics Section*
|
Abstract - #301200 |
Title:
|
Allometric Extension for Multivariate Regression Models
|
Author(s):
|
Thaddeus Tarpey*+ and Christopher Ivey
|
Affiliation(s):
|
Wright State University and Blandy Experimental Farm
|
Address:
|
120 MM Building, Dayton, Ohio, 45435, USA
|
Keywords:
|
principal components ; reduced rank regression ; multivariate regression
|
Abstract:
|
The two-sample allometric extension model is extended to the multivariate regression setting. In morphometric studies, where the first principal component corresponds to a size variable, the allometric extension model implies that regressor variables under consideration effect only the size of the plant and not its shape. A study of plant growth is illustrated where leaf measurements are regressed on soil variables, including contaminants. Departures from the allometric extension model imply that the contaminants effect not only the size of the leaf but also its shape. The allometric extension model for multivariate regression also provides a useful interpretation to the reduced rank regression model of rank equal to one.
|