|
Activity Number:
|
258
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
International Society of Bayesian Analysis
|
| Abstract - #307915 |
|
Title:
|
Bayesian Nonproportional Hazards Survival Analysis
|
|
Author(s):
|
Wesley Johnson*+
|
|
Companies:
|
University of California, Irvine
|
|
Address:
|
Department of Statistics, Irvine, CA, 92697,
|
|
Keywords:
|
Nonparametric ; Gibbs Sampling ; Dependent Dirichlet Process
|
|
Abstract:
|
We discuss some recent advances in nonparametric approaches to survival data. In particular we discuss a model that allows survival curves to cross, possibly due to a toxic treatment effect resulting in early deaths but which is ultimately effective as a treatment. An illustration based on a cancer clinical trial is given. As time permits, we also discuss some issues related to model selection for survival data with time dependent covariates.
|
