|
Activity Number:
|
382
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Risk Analysis
|
| Abstract - #305022 |
|
Title:
|
Cox Proportional Hazards Model with Brownian-Like Predictor
|
|
Author(s):
|
Yulei Zhang*+ and Ian McKeague
|
|
Companies:
|
Columbia University and Columbia University
|
|
Address:
|
722 West 168th street, 6th floor, Department of Biostatistics, New York, NY, 10032,
|
|
Keywords:
|
Cox model ; functional data ; longitudinal data ; M-estimation ; emprical processes
|
|
Abstract:
|
We investigate the Cox proportional hazards model involving covariate whose paths behave like Brownian Motion. The parameter of interest is the unknown time point at which subjects' covariate values are most predictive of their survival times. We adopted the M-estimator framework in van der Vaart and Wellner (1996), derived the convergence rate and asymptotic distribution of the estimator of the most predictive time point. Its asymptotic distribution is the maximizer of a 2-sided Gaussian process with a negative drift. This work falls into the areas of risk analysis, functional data analysis, joint modeling of (some specific) longitudinal data and survival data. The main tools used for proof is Empirical Processes theory. This is a joint work with Ian McKeague.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
