A weighted least-squares cross-validation bandwidth selector for kernel density estimation. Issue 7 (3rd April 2017)

• Record Type: Journal Article
• Authors: Tenreiro, C.
• Is Part Of: Communications in statistics. Volume 46:Issue 7(2017)
• Page Start: 3438
• Page End: 3458
• Publication Date: 2017-04-03
• Subjects: Kernel density estimation -- Bandwidth selection -- Cross-validation; Primary 62G07 -- Secondary 62G20; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2015.1062108
• Languages: English
• ISSNs: 0361-0926

Abstract

ABSTRACT: Since the late 1980s, several methods have been considered in the literature to reduce the sample variability of the least-squares cross-validation bandwidth selector for kernel density estimation. In this article, a weighted version of this classical method is proposed and its asymptotic and finite-sample behavior is studied. The simulation results attest that the weighted cross-validation bandwidth performs quite well, presenting a better finite-sample performance than the standard cross-validation method for "easy-to-estimate" densities, and retaining the good finite-sample performance of the standard cross-validation method for "hard-to-estimate" ones.