Sufficient variable screening via directional regression with censored response

TL;DR

This paper introduces a model-free directional regression method for ultrahigh dimensional variable screening with censored responses, ensuring sure screening properties and demonstrating effectiveness through simulations and real data.

Contribution

It presents a novel, model-free directional regression approach for variable screening with censored data, including iterative and stability algorithms.

Findings

The method achieves sure screening property under exponential divergence of p.

It performs well in simulations and real data analysis.

The approach is adaptable to complex models.

Abstract

We in this paper propose a directional regression based approach for ultrahigh dimensional sufficient variable screening with censored responses. The new method is designed in a model-free manner and thus can be adapted to various complex model structures. Under some commonly used assumptions, we show that the proposed method enjoys the sure screening property when the dimension p diverges at an exponential rate of the sample size n. To improve the marginal screening method, the corresponding iterative screening algorithm and stability screening algorithm are further equipped. We demonstrate the effectiveness of the proposed method through simulation studies and a real data analysis.