|
Activity Number:
|
525
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304757 |
|
Title:
|
Model Averaging for Multifactor Stability Studies
|
|
Author(s):
|
Seth Clark*+ and Robert Noble
|
|
Companies:
|
Merck & Co., Inc. and Miami University of Ohio
|
|
Address:
|
899 Shamrock Ave, State College, PA, 16801,
|
|
Keywords:
|
bayesian model averaging ; analysis of covariance ; model selection ; information criterion
|
|
Abstract:
|
Typical model selection ignores a variance component associated with model uncertainty that sets up "discontinuities" in the inferences across different data sets. For stability data in particular, there are discontinuities in predictions of stability profiles, degrees-of-freedom, shelf-life estimation, and the scientific interpretation of the fitted stability study models. We show examples that illustrate even very small changes in the data can lead to drastically different models and shelf-life estimates. We show how model averaging can address these shortcomings.
|
- 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 |
