|
Activity Number:
|
404
- Novel Methods for High-Dimensional and Large-Scale Survival Data
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 1:00 PM to 2:50 PM
|
|
Sponsor:
|
WNAR
|
|
Abstract #311157
|
|
|
Title:
|
Double-Slicing Assisted Sufficient Dimension Reduction for High-Dimensional Censored Data
|
|
Author(s):
|
Shanshan Ding* and Wei Qian and Lan Wang
|
|
Companies:
|
University of Delaware and University of Delaware and University of Minnesota
|
|
Keywords:
|
censored data;
dimension reduction;
variable selection;
ultrahigh dimension;
nonparametric estimation
|
|
Abstract:
|
This talk presents a unified framework and an efficient algorithm for analyzing high-dimensional survival data under weak modeling assumptions. In particular, it imposes neither parametric distributional assumption nor linear regression assumption. It only assumes that the survival time depends on a high-dimensional covariate vector through low-dimensional linear combinations of covariates. The censoring time is allowed to be conditionally independent of the survival time given the covariates. This general framework includes many popular parametric and semiparametric survival regression models as special cases. The proposed algorithm produces a number of practically useful outputs with theoretical guarantees, including a consistent estimate of the sufficient dimension reduction subspace, a uniformly consistent Kaplan-Meier type estimator of the conditional distribution function of the survival time and a consistent estimator of the conditional quantile survival time. Our asymptotic results significantly extend the classical theory of sufficient dimension reduction for censored data and the celebrated nonparametric Kaplan-Meier estimator to the ultrahigh dimensional setting.
|
Authors who are presenting talks have a * after their name.
