|
Activity Number:
|
40
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
SSC
|
| Abstract - #307933 |
|
Title:
|
A Class of Simple Lifetime Regression Models Resulting from First-Hitting Times
|
|
Author(s):
|
Thierry Duchesne*+
|
|
Companies:
|
Laval University
|
|
Address:
|
Departement de mathematiques & statistique, Quebec, QC, G1K 7P4, Canada
|
|
Keywords:
|
lifetime regression ; threshold regression ; collapsible model ; time-varying covariate
|
|
Abstract:
|
Oakes (1995), Kordonsky and Gertsbakh (1997) and Duchesne and Lawless (2000) consider a class of simple regression models for lifetime given the history of an increasing covariate that can be viewed as an alternate measure of age. These models are appealing due to their interpretability and potential for non-parametric inference, and they fit several types of datasets quite well; Oakes (1995) uses them to model the lifetime of miners exposed to asbestos dust and in an analysis of the Channing House retirement center data, whereas Kordonsky and Gertsbakh (1997) and Duchesne and Lawless (2000) use them in reliability data analyses. In this talk, we will see how these models can be obtained as the distribution of the first time that a wear (or health) process that is driven by usage (or exposure) hits a threshold. Potential for specific applications will also be discussed.
|
- 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 2007 program |
