|
Activity Number:
|
467
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #303235 |
|
Title:
|
The Orthogonal Interactions Model for Unreplicated Factorial Experiments
|
|
Author(s):
|
Clayton A. Barker*+ and Leonard A. Stefanski and Jason Osborne
|
|
Companies:
|
SAS Institute Inc. and North Carolina State University and North Carolina State University
|
|
Address:
|
100 SAS Campus Dr, , Cary, NC, 27513,
|
|
Keywords:
|
Unreplicated factorial experiments ; Estimated degrees of freedom ; AMMI Model
|
|
Abstract:
|
Unreplicated factorial experiments do not allow for estimation of the interaction terms in full factorial models, thus limiting the scope of inference. Existing methods for such data (e.g., Tukey's one-degree-of-freedom model, linear-bilinear models, and the additive main effects and multiplicative interactions (AMMI) model) enable error-variance estimation by constraining interactions to have a multiplicative form. We develop a new constrained-interaction model by forcing interactions and main effects to be orthogonal. Because its interactions are not multiplicative, the new orthogonal-interactions (OI) model complements existing methods based on multiplicative interactions. Like the AMMI model, the OI model sometimes requires estimated degrees of freedom. We present a new method for estimating degrees of freedom whose utility transcends the OI model.
|
- 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 |
