|
Activity Number:
|
103
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #309873 |
|
Title:
|
Test of Fit to a Semiparametric Bivariate Family with Censored Data
|
|
Author(s):
|
Shulamith Gross*+ and Catherine Huber Carol
|
|
Companies:
|
Baruch College/CUNY and Université Paris V
|
|
Address:
|
One Baruch Way, New York, NY, 10010,
|
|
Keywords:
|
bivariate ; test of fit ; censoring ; semiparametric
|
|
Abstract:
|
In this talk I shall introduce a test of fit to a semiparametric family of Bivariate distributions recently introduced by Gross and Huber. The general family is multivariate and is particularly appropriate for reliability applications in a multi-component system, or in dependent competing risk problems. The family was proposed as a possible alternative to the semiparametric copula models or frailty models for multivariate data, and adapts readily to applications with censored data. The test is distribution free in the uncensored case. The power and asymptotic validity of the proposed test of fit will be studied via simulations. The talk is based on joint work with Catherine Huber.
|
