High-Dimensional Quantile Regression: Convolution Smoothing and Concave Regularization

TL;DR

This paper introduces a convolution-smoothed quantile regression method with iteratively reweighted regularization, achieving optimal convergence rates and oracle properties in high-dimensional heterogenous data analysis.

Contribution

It develops a novel smoothed quantile regression approach that overcomes non-smoothness issues, providing theoretical guarantees and improved estimation accuracy.

Findings

Achieves optimal convergence rates for high-dimensional quantile regression.

Establishes oracle properties under minimal signal strength conditions.

Numerical studies validate theoretical advantages.

Abstract

$\ell_1$-penalized quantile regression is widely used for analyzing high-dimensional data with heterogeneity. It is now recognized that the $\ell_1$-penalty introduces non-negligible estimation bias, while a proper use of concave regularization may lead to estimators with refined convergence rates and oracle properties as the signal strengthens. Although folded concave penalized $M$-estimation with strongly convex loss functions have been well studied, the extant literature on quantile regression is relatively silent. The main difficulty is that the quantile loss is piecewise linear: it is non-smooth and has curvature concentrated at a single point. To overcome the lack of smoothness and strong convexity, we propose and study a convolution-type smoothed quantile regression with iteratively reweighted $\ell_1$-regularization. The resulting smoothed empirical loss is twice continuously differentiable and (provably) locally strongly convex with high probability. We show that the iteratively reweighted $\ell_1$-penalized smoothed quantile regression estimator, after a few iterations, achieves the optimal rate of convergence, and moreover, the oracle rate and the strong oracle property under an almost necessary and sufficient minimum signal strength condition. Extensive numerical studies corroborate our theoretical results.