Finding Dantzig selectors with a proximity operator based fixed-point algorithm

TL;DR

This paper introduces a fixed-point algorithm based on proximity operators to efficiently compute the Dantzig selector in linear regression, showing comparable accuracy to existing methods but with faster convergence.

Contribution

The paper proposes a novel fixed-point iterative method for finding the Dantzig selector, improving computational speed over traditional approaches.

Findings

The proposed method achieves similar accuracy to the alternating direction method.

Numerical simulations show the new method converges faster.

The method performs well on both synthetic and real datasets.

Abstract

In this paper, we study a simple iterative method for finding the Dantzig selector, which was designed for linear regression problems. The method consists of two main stages. The first stage is to approximate the Dantzig selector through a fixed-point formulation of solutions to the Dantzig selector problem. The second stage is to construct a new estimator by regressing data onto the support of the approximated Dantzig selector. We compare our method to an alternating direction method, and present the results of numerical simulations using both the proposed method and the alternating direction method on synthetic and real data sets. The numerical simulations demonstrate that the two methods produce results of similar quality, however the proposed method tends to be significantly faster.