Spearman Rank Correlation Screening for Ultrahigh-dimensional Censored Data

TL;DR

This paper introduces a model-free Spearman rank correlation screening method for ultrahigh-dimensional censored data, demonstrating robustness and superior performance in various challenging scenarios including heavy tails and high censoring.

Contribution

The paper proposes a novel Spearman rank correlation based screening procedure that is robust, model-free, and effective for ultrahigh-dimensional censored data, with proven sure-screening properties.

Findings

Performs well with heavy-tailed distributions

Effective under high censoring rates

Outperforms existing nonparametric screening methods

Abstract

Herein, we propose a Spearman rank correlation based screening procedure for ultrahigh-dimensional data with censored response case. The proposed method is model-free without specifying any regression forms of predictors or response variable and is robust under the unknown monotone transformations of these response variable and predictors. The sure-screening and rank-consistency properties are established under some mild regularity conditions. Simulation studies demonstrate that the new screening method performs well in the presence of a heavy-tailed distribution, strongly dependent predictors or outliers and that offers superior performance over the existing nonparametric screening procedures. In particular, the new screening method still works well when a response variable is observed under a high censoring rate. An illustrative example is provided.