|
Activity Number:
|
135
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #309220 |
|
Title:
|
Laplace's Approximation for Relative Risk Frailty Models
|
|
Author(s):
|
Shibao Feng*+ and Lei Nie and Robert A. Wolfe
|
|
Companies:
|
Genentech, Inc. and Georgetown University Medical Center and University Renal Research and Education Association
|
|
Address:
|
1 DNA Way MS 441B, South San Francisco, CA, 94080-4990,
|
|
Keywords:
|
Multivariate survival analysis ; Mixed Poisson regression models ; Empirical Bayes estimates
|
|
Abstract:
|
Relative risk frailty models are used extensively in analyzing clustered and/or recurrent time-to-event data. In this paper, Laplace's approximation for integrals is applied to marginal distributions of data arising from parametric (e.g. piecewise exponential) relative risk frailty models. A full likelihood approach (Feng, Wolfe, and Port 2005) is used to estimate the parameters. Under regularity conditions, the full likelihood estimators are shown to be consistent with a rate of convergence depending on both the number of subjects and number of observations per subject. We compare the full likelihood estimators against alternative estimators using limited simulation and demonstrate the utility of the full likelihood approach by analyzing U.S. patient waiting time to deceased kidney transplant data.
|
