High-Dimensional Quantile Regression: Convolution Smoothing and Concave Regularization

09/12/2021
by   Kean Ming Tan, et al.
0

ℓ_1-penalized quantile regression is widely used for analyzing high-dimensional data with heterogeneity. It is now recognized that the ℓ_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 ℓ_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 ℓ_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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/11/2022

Retire: Robust Expectile Regression in High Dimensions

High-dimensional data can often display heterogeneity due to heterosceda...
research
08/21/2022

High-Dimensional Composite Quantile Regression: Optimal Statistical Guarantees and Fast Algorithms

The composite quantile regression (CQR) was introduced by Zou and Yuan [...
research
05/05/2022

A Unified Algorithm for Penalized Convolution Smoothed Quantile Regression

Penalized quantile regression (QR) is widely used for studying the relat...
research
07/06/2021

A provable two-stage algorithm for penalized hazards regression

From an optimizer's perspective, achieving the global optimum for a gene...
research
09/04/2023

Smoothing ADMM for Sparse-Penalized Quantile Regression with Non-Convex Penalties

This paper investigates quantile regression in the presence of non-conve...
research
05/21/2012

Variance function estimation in high-dimensions

We consider the high-dimensional heteroscedastic regression model, where...
research
05/11/2016

Asymptotic properties for combined L_1 and concave regularization

Two important goals of high-dimensional modeling are prediction and vari...

Please sign up or login with your details

Forgot password? Click here to reset