|
Activity Number:
|
578
- Survival Analysis and Semiparametic and Nonparametric Models
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 2, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Biometrics Section
|
|
Abstract #323624
|
|
|
Title:
|
Semiparametric Single Index Regression Models for Censored Outcomes
|
|
Author(s):
|
Jin Wang* and Donglin Zeng and Danyu Lin
|
|
Companies:
|
UNC Chapel Hill and University of North Carolina and University of North Carolina
|
|
Keywords:
|
survival analysis ;
B-splines ;
nonparametric maximum likelihood
|
|
Abstract:
|
We propose a flexible single-index model to allow complex treatment-covariate interactions and to derive a simple linear treatment rule for personalized treatment decision. Our model extends the proportional hazards models by using a monotone regression function. The inference procedures are based on sieve estimation that includes single-index parameters in the basis expansion. We obtain the theoretical results by deriving the necessary rates of convergence for the nonparamatric estimator of the arbitrary regression function. We provide simultaneous inference on single-index parameters and other regression parameters. Simulation studies are conducted to assess the finite sample performance. An application to a multiple-type cancer study is presented to illustrate our methods.
|
Authors who are presenting talks have a * after their name.
