Boundary performance of the beta kernel estimators

Shunpu Zhang, Rohana J. Karunamuni

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

The beta kernel estimators are shown in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] to be non-negative and have less severe boundary problems than the conventional kernel estimator. Numerical results in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] further show that beta kernel estimators have better finite sample performance than some of the widely used boundary corrected estimators. However, our study finds that the numerical comparisons of Chen are confounded with the choice of the bandwidths and the quantities being compared. In this paper, we show that the performances of the beta kernel estimators are very similar to that of the reflection estimator, which does not have the boundary problem only for densities exhibiting a shoulder at the endpoints of the support. For densities not exhibiting a shoulder, we show that the beta kernel estimators have a serious boundary problem and their performances at the boundary are inferior to that of the well-known boundary kernel estimator.

Original languageEnglish (US)
Pages (from-to)81-104
Number of pages24
JournalJournal of Nonparametric Statistics
Volume22
Issue number1
DOIs
StatePublished - Jan 1 2010

Fingerprint

Kernel Estimator
Boundary Problem
Density Function
Estimator
Kernel estimator
Numerical Comparisons
Non-negative
Bandwidth
Numerical Results

Keywords

  • Beta kernel estimator
  • Boundary kernel estimator
  • Boundary problem
  • Nonparametric density estimation
  • Reflection estimator

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Boundary performance of the beta kernel estimators. / Zhang, Shunpu; Karunamuni, Rohana J.

In: Journal of Nonparametric Statistics, Vol. 22, No. 1, 01.01.2010, p. 81-104.

Research output: Contribution to journalArticle

Zhang, Shunpu ; Karunamuni, Rohana J. / Boundary performance of the beta kernel estimators. In: Journal of Nonparametric Statistics. 2010 ; Vol. 22, No. 1. pp. 81-104.
@article{1141dedc50fc4d8083a77e8f9885a9f4,
title = "Boundary performance of the beta kernel estimators",
abstract = "The beta kernel estimators are shown in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] to be non-negative and have less severe boundary problems than the conventional kernel estimator. Numerical results in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] further show that beta kernel estimators have better finite sample performance than some of the widely used boundary corrected estimators. However, our study finds that the numerical comparisons of Chen are confounded with the choice of the bandwidths and the quantities being compared. In this paper, we show that the performances of the beta kernel estimators are very similar to that of the reflection estimator, which does not have the boundary problem only for densities exhibiting a shoulder at the endpoints of the support. For densities not exhibiting a shoulder, we show that the beta kernel estimators have a serious boundary problem and their performances at the boundary are inferior to that of the well-known boundary kernel estimator.",
keywords = "Beta kernel estimator, Boundary kernel estimator, Boundary problem, Nonparametric density estimation, Reflection estimator",
author = "Shunpu Zhang and Karunamuni, {Rohana J.}",
year = "2010",
month = "1",
day = "1",
doi = "10.1080/10485250903124984",
language = "English (US)",
volume = "22",
pages = "81--104",
journal = "Journal of Nonparametric Statistics",
issn = "1048-5252",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

TY - JOUR

T1 - Boundary performance of the beta kernel estimators

AU - Zhang, Shunpu

AU - Karunamuni, Rohana J.

PY - 2010/1/1

Y1 - 2010/1/1

N2 - The beta kernel estimators are shown in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] to be non-negative and have less severe boundary problems than the conventional kernel estimator. Numerical results in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] further show that beta kernel estimators have better finite sample performance than some of the widely used boundary corrected estimators. However, our study finds that the numerical comparisons of Chen are confounded with the choice of the bandwidths and the quantities being compared. In this paper, we show that the performances of the beta kernel estimators are very similar to that of the reflection estimator, which does not have the boundary problem only for densities exhibiting a shoulder at the endpoints of the support. For densities not exhibiting a shoulder, we show that the beta kernel estimators have a serious boundary problem and their performances at the boundary are inferior to that of the well-known boundary kernel estimator.

AB - The beta kernel estimators are shown in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] to be non-negative and have less severe boundary problems than the conventional kernel estimator. Numerical results in Chen [S.X. Chen, Beta kernel estimators for density functions, Comput. Statist. Data Anal. 31 (1999), pp. 131-145] further show that beta kernel estimators have better finite sample performance than some of the widely used boundary corrected estimators. However, our study finds that the numerical comparisons of Chen are confounded with the choice of the bandwidths and the quantities being compared. In this paper, we show that the performances of the beta kernel estimators are very similar to that of the reflection estimator, which does not have the boundary problem only for densities exhibiting a shoulder at the endpoints of the support. For densities not exhibiting a shoulder, we show that the beta kernel estimators have a serious boundary problem and their performances at the boundary are inferior to that of the well-known boundary kernel estimator.

KW - Beta kernel estimator

KW - Boundary kernel estimator

KW - Boundary problem

KW - Nonparametric density estimation

KW - Reflection estimator

UR - http://www.scopus.com/inward/record.url?scp=77049101016&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77049101016&partnerID=8YFLogxK

U2 - 10.1080/10485250903124984

DO - 10.1080/10485250903124984

M3 - Article

AN - SCOPUS:77049101016

VL - 22

SP - 81

EP - 104

JO - Journal of Nonparametric Statistics

JF - Journal of Nonparametric Statistics

SN - 1048-5252

IS - 1

ER -