|
Activity Number:
|
407
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #304997 |
|
Title:
|
A Unified Competing Risks Cure Rate Model for Failure Data
|
|
Author(s):
|
Sanjib Basu*+
|
|
Companies:
|
Northern Illinois University
|
|
Address:
|
DU 366, Division of Statistics, De Kalb, IL, 60564,
|
|
Keywords:
|
Competing risks ; cure rate ; identifiability ; model selection
|
|
Abstract:
|
The competing risks approach considers the setting where a unit or subject is exposed to multiple risks simultaneously. A Cure rate or limited failure model, on the other hand, postulates a fraction of units or subjects to be cured or failure-free. We describe a model that unifies the competing risks and cure rate approaches. We discuss identifiability issues of the proposed model, compare it with two different restricted model formulations and conclude with a detailed application.
|
