An empirical study of indirect cross–validation

Olga Savchuk; Jeffrey Hart; Simon Sheather

doi:10.1142/9789814340564_0017

An empirical study of indirect cross–validation

Olga Savchuk, Jeffrey Hart, Simon Sheather

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

4 Scopus citations

Abstract

In this paper we provide insight into the empirical properties of indirect cross-validation (ICV), a new method of bandwidth selection for kernel density estimators. First, we describe the method and report on the theoretical results used to develop a practical-purpose model for certain ICV parameters. Next, we provide a detailed description of a numerical study that shows that the ICV method usually outperforms least squares cross-validation (LSCV) in finite samples. One of the major advantages of ICV is its increased stability compared to LSCV. Two real data examples show the benefit of using both ICV and a local version of ICV.

Original language	English
Title of host publication	Nonparametric Statistics and Mixture Models
Subtitle of host publication	A Festschrift in Honor of Thomas P Hettmansperger
Pages	288-308
Number of pages	21
ISBN (Electronic)	9789814340564
DOIs	https://doi.org/10.1142/9789814340564_0017
State	Published - Jan 1 2011

Bibliographical note

Publisher Copyright:
© 2011 by World Scientific Publishing Co. Pte. Ltd. All rights reserved.

Keywords

Bandwidth selection
Cross-validation
Integrated squared error
Kernel density estimation
Mean integrated squared error

ASJC Scopus subject areas

General Mathematics

Access to Document

10.1142/9789814340564_0017

Cite this

@inbook{b8c48805c90a46a2b9dfbd444d2ca47b,

title = "An empirical study of indirect cross–validation",

abstract = "In this paper we provide insight into the empirical properties of indirect cross-validation (ICV), a new method of bandwidth selection for kernel density estimators. First, we describe the method and report on the theoretical results used to develop a practical-purpose model for certain ICV parameters. Next, we provide a detailed description of a numerical study that shows that the ICV method usually outperforms least squares cross-validation (LSCV) in finite samples. One of the major advantages of ICV is its increased stability compared to LSCV. Two real data examples show the benefit of using both ICV and a local version of ICV.",

keywords = "Bandwidth selection, Cross-validation, Integrated squared error, Kernel density estimation, Mean integrated squared error",

author = "Olga Savchuk and Jeffrey Hart and Simon Sheather",

year = "2011",

month = jan,

day = "1",

doi = "10.1142/9789814340564_0017",

language = "English",

isbn = "9814340553",

pages = "288--308",

booktitle = "Nonparametric Statistics and Mixture Models",

}

TY - CHAP

T1 - An empirical study of indirect cross–validation

AU - Savchuk, Olga

AU - Hart, Jeffrey

AU - Sheather, Simon

PY - 2011/1/1

Y1 - 2011/1/1

N2 - In this paper we provide insight into the empirical properties of indirect cross-validation (ICV), a new method of bandwidth selection for kernel density estimators. First, we describe the method and report on the theoretical results used to develop a practical-purpose model for certain ICV parameters. Next, we provide a detailed description of a numerical study that shows that the ICV method usually outperforms least squares cross-validation (LSCV) in finite samples. One of the major advantages of ICV is its increased stability compared to LSCV. Two real data examples show the benefit of using both ICV and a local version of ICV.

AB - In this paper we provide insight into the empirical properties of indirect cross-validation (ICV), a new method of bandwidth selection for kernel density estimators. First, we describe the method and report on the theoretical results used to develop a practical-purpose model for certain ICV parameters. Next, we provide a detailed description of a numerical study that shows that the ICV method usually outperforms least squares cross-validation (LSCV) in finite samples. One of the major advantages of ICV is its increased stability compared to LSCV. Two real data examples show the benefit of using both ICV and a local version of ICV.

KW - Bandwidth selection

KW - Cross-validation

KW - Integrated squared error

KW - Kernel density estimation

KW - Mean integrated squared error

UR - http://www.scopus.com/inward/record.url?scp=84973136612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84973136612&partnerID=8YFLogxK

U2 - 10.1142/9789814340564_0017

DO - 10.1142/9789814340564_0017

M3 - Chapter

AN - SCOPUS:84973136612

SN - 9814340553

SN - 9789814340557

SP - 288

EP - 308

BT - Nonparametric Statistics and Mixture Models

ER -

An empirical study of indirect cross–validation

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this