Adaptive Test for High Dimensional Quantile Regression

TL;DR

This paper introduces an adaptive testing procedure for high-dimensional quantile regression coefficients that combines max-type and sum-type tests using a Cauchy combination, providing robustness across different sparsity patterns.

Contribution

It establishes the asymptotic independence of two test statistics and proposes a novel Cauchy combination test that enhances power and size control in high-dimensional settings.

Findings

The proposed test outperforms existing methods in simulations.

It maintains robustness across various sparsity levels.

Demonstrates effectiveness on real data applications.

Abstract

Testing high-dimensional quantile regression coefficients is crucial, as tail quantiles often reveal more than the mean in many practical applications. Nevertheless, the sparsity pattern of the alternative hypothesis is typically unknown in practice, posing a major challenge. To address this, we propose an adaptive test that remains powerful across both sparse and dense alternatives.We first establish the asymptotic independence between the max-type test statistic proposed by \citet{tang2022conditional} and the sum-type test statistic introduced by \citet{chen2024hypothesis}. Building on this result, we propose a Cauchy combination test that effectively integrates the strengths of both statistics and achieves robust performance across a wide range of sparsity levels. Simulation studies and real data applications demonstrate that our proposed procedure outperforms existing methods in terms of both size control and power.