Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 485 - A Unified View on Model Selection
Type: Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
Sponsor: Biometrics Section
Abstract #313258
Title: A Splitting and Smoothing Approach for the Cox Regression with High-Dimensional Interval Censored Data
Author(s): Rui Yang* and Gang Li
Companies: UCLA/Covance Inc. and University of California, Los Angeles
Keywords: Interval Censoring; Cox Regression; High Dimensional Variable Selection; Survival Analysis; Oracle Property; Semi-parametric Inference

This paper proposes a new framework of estimation and inferences for the Cox proportional hazards model with high dimensional interval-censored data. By smoothing over partial regression estimates based on a given variable selection scheme such as adaptive lasso, we reduce the problem to a low-dimensional maximum-likelihood estimation. The procedure utilizes data splitting along with variable selection and partial regression. The method is proven to enjoy the desirable oracle property. An extensive simulation study is conducted and indicates that the proposed approach works well in practical situations and deals with the collinearity problem better than the other oracle-like methods. A real-world application is also provided.  

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program