Premium
Generalized least squares cross‐validation in kernel density estimation
Author(s) -
Zhang Jin
Publication year - 2015
Publication title -
statistica neerlandica
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.52
H-Index - 39
eISSN - 1467-9574
pISSN - 0039-0402
DOI - 10.1111/stan.12061
Subject(s) - cross validation , kernel density estimation , bandwidth (computing) , kernel (algebra) , variable kernel density estimation , multivariate kernel density estimation , mathematics , generalized least squares , algorithm , density estimation , computer science , least squares function approximation , statistics , kernel method , mathematical optimization , artificial intelligence , support vector machine , discrete mathematics , telecommunications , estimator
The kernel density estimation is a popular method in density estimation. The main issue is bandwidth selection, which is a well‐known topic and is still frustrating statisticians. A robust least squares cross‐validation bandwidth is proposed, which significantly improves the classical least squares cross‐validation bandwidth for its variability and undersmoothing, adapts to different kinds of densities, and outperforms the existing bandwidths in statistical literature and software.