Premium
Combining local and global smoothing in multivariate density estimation
Author(s) -
Azzalini Adelchi
Publication year - 2016
Publication title -
stat
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 18
ISSN - 2049-1573
DOI - 10.1002/sta4.129
Subject(s) - smoothing , multivariate statistics , density estimation , multivariate kernel density estimation , computer science , estimation , parametric statistics , cluster analysis , econometrics , mathematical optimization , mathematics , artificial intelligence , machine learning , statistics , engineering , computer vision , kernel method , estimator , variable kernel density estimation , support vector machine , systems engineering
Non‐parametric estimation of a multivariate density is tackled via a method that combines traditional local smoothing with a form of global smoothing but without imposing a rigid structure. Simulation work delivers encouraging indications on the effectiveness of the method. An application to density‐based clustering illustrates a possible usage. Copyright © 2016 John Wiley & Sons, Ltd.