|
Activity Number:
|
64
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304599 |
|
Title:
|
Frailty Models with Nonparametric Additive Covariate Functions
|
|
Author(s):
|
Zhangsheng Yu*+ and Xihong Lin
|
|
Companies:
|
Indiana University School of Medicine and Harvard School of Public Health
|
|
Address:
|
410 W. 10th St Suite 3000, Indianapolis, IN, 46202,
|
|
Keywords:
|
survival analysis ; smoothing spline ; penallized likelihood
|
|
Abstract:
|
We propose a double penalized partial likelihood (DPPL) for nonparametric regression of Gaussian frailty models with additive covariate functions. This provides a unified approach for covariate function and smoothing parameter estimation. Numerical analysis shows that maximum likelihood method for estimating smoothing parameters works well. The point estimator and standard error estimation of covariate functions perform well. With this unified approach, the nonparametric/semiparametric regression of frailty model can take advantage of the increasing popularity of parametric frailty model in practice as the software package for frailty become available in the future. We also apply the proposed method to western Kenya parasitemia study.
|
