|
Activity Number:
|
458
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #306543 |
|
Title:
|
A Bayesian Dynamic Frailty Model for Recurrent Events
|
|
Author(s):
|
Changhong Song and Lynn Kuo*+
|
|
Companies:
|
University of Connecticut and University of Connecticut
|
|
Address:
|
Department of Statistics, Storrs, CT, 06269,
|
|
Keywords:
|
|
|
Abstract:
|
In many medical studies, times to occurrence of a specific event---such as recurring hospitalizations or tumors---have been collected. To analyze this kind of data, within subject association must be modeled to ensure correct inferences for the treatment effects. Some dynamic frailty models have been proposed to model this association and the evolution of individual effect over time. In this study, we propose a new family of dynamic frailty model to better describe each individual's risk that changes with age during the trial. In the new model, the individual frailty effect is modeled as a time-varying effect with unknown change points. Both the unobserved change points and the intensities of the frailty function are modeled as latent variables in the model. The implementation of Bayesian inference and model selection using a Markov chain Monte Carlo (MCMC) algorithm is developed.
|
