Variable selection in additive quantile regression using nonconcave penalty. Issue 6 (1st November 2016)

• Record Type: Journal Article
• Authors: Zhao, Kaifeng; Lian, Heng
• Is Part Of: Statistics. Volume 50:Issue 6(2016)
• Page Start: 1276
• Page End: 1289
• Publication Date: 2016-11-01
• Subjects: Additive models -- oracle property -- SCAD penalty -- schwartz-type information criterion; Mathematical statistics -- Periodicals; 519.505
• DOI: 10.1080/02331888.2016.1221954
• Languages: English
• ISSNs: 0233-1888

Abstract

ABSTRACT: This paper considers variable selection in additive quantile regression based on group smoothly clipped absolute deviation (gSCAD) penalty. Although shrinkage variable selection in additive models with least-squares loss has been well studied, quantile regression is sufficiently different from mean regression to deserve a separate treatment. It is shown that the gSCAD estimator can correctly identify the significant components and at the same time maintain the usual convergence rates in estimation. Simulation studies are used to illustrate our method.