|
Activity Number:
|
6
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 29, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #307844 |
|
Title:
|
Asymptotically Optimal Tests Under Loss of Identifiability in Semiparametric Models
|
|
Author(s):
|
Rui Song and Michael Kosorok*+ and Jason Fine
|
|
Companies:
|
The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill and University of Wisconsin-Madison
|
|
Address:
|
Department of Biostatistics , Chapel Hill , NC, 27599-7420,
|
|
Keywords:
|
Semiparametric methods ; Optimal tests ; Loss of identifiability ; Change-point models ; Transformation models ; Mixture models
|
|
Abstract:
|
We consider tests of hypotheses when the parameters are not identifiable under the null in semiparametric models. Under a weighted average power criterion, exponential average tests are characterized and shown to be asymptotically optimal. The results can be applied to a variety of semiparametric models, for example, tests of presence of change-point in transformation models, tests of regression parameters in gamma frailty models, and tests of the number of mixture components in finite components mixture models are discussed. We also propose a modified weighted bootstrap for computing the critical values of the test statistic.
|
