On piecewise linear density estimators

J. Beirlant, A. Berlinet, L. Györfi

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We study piecewise linear density estimators from the L1 point of view: the frequency polygons investigated by SCOTT (1985) and JONES et al. (1997), and a new piecewise linear histogram. In contrast to the earlier proposals, a unique multivariate generalization of the new piecewise linear histogram is available. All these estimators are shown to be universally L1 strongly consistent. We derive large deviation inequalities. For twice differentiable densities with compact support their expected L1 error is shown to have the same rate of convergence as have kernel density estimators. Some simulated examples are presented.

Original languageEnglish
Pages (from-to)287-308
Number of pages22
JournalStatistica Neerlandica
Volume53
Issue number3
Publication statusPublished - Nov 1999

Fingerprint

Linear Estimator
Density Estimator
Piecewise Linear
Histogram
Deviation Inequalities
Kernel Density Estimator
Compact Support
Large Deviations
Polygon
Differentiable
Rate of Convergence
Estimator
Linear density

Keywords

  • Asymptotics
  • Histogram
  • Nonparametric density estimation

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Beirlant, J., Berlinet, A., & Györfi, L. (1999). On piecewise linear density estimators. Statistica Neerlandica, 53(3), 287-308.

On piecewise linear density estimators. / Beirlant, J.; Berlinet, A.; Györfi, L.

In: Statistica Neerlandica, Vol. 53, No. 3, 11.1999, p. 287-308.

Research output: Contribution to journalArticle

Beirlant, J, Berlinet, A & Györfi, L 1999, 'On piecewise linear density estimators', Statistica Neerlandica, vol. 53, no. 3, pp. 287-308.
Beirlant J, Berlinet A, Györfi L. On piecewise linear density estimators. Statistica Neerlandica. 1999 Nov;53(3):287-308.
Beirlant, J. ; Berlinet, A. ; Györfi, L. / On piecewise linear density estimators. In: Statistica Neerlandica. 1999 ; Vol. 53, No. 3. pp. 287-308.
@article{e4a7fa87728f4a3e8dacf5eb1398844b,
title = "On piecewise linear density estimators",
abstract = "We study piecewise linear density estimators from the L1 point of view: the frequency polygons investigated by SCOTT (1985) and JONES et al. (1997), and a new piecewise linear histogram. In contrast to the earlier proposals, a unique multivariate generalization of the new piecewise linear histogram is available. All these estimators are shown to be universally L1 strongly consistent. We derive large deviation inequalities. For twice differentiable densities with compact support their expected L1 error is shown to have the same rate of convergence as have kernel density estimators. Some simulated examples are presented.",
keywords = "Asymptotics, Histogram, Nonparametric density estimation",
author = "J. Beirlant and A. Berlinet and L. Gy{\"o}rfi",
year = "1999",
month = "11",
language = "English",
volume = "53",
pages = "287--308",
journal = "Statistica Neerlandica",
issn = "0039-0402",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - On piecewise linear density estimators

AU - Beirlant, J.

AU - Berlinet, A.

AU - Györfi, L.

PY - 1999/11

Y1 - 1999/11

N2 - We study piecewise linear density estimators from the L1 point of view: the frequency polygons investigated by SCOTT (1985) and JONES et al. (1997), and a new piecewise linear histogram. In contrast to the earlier proposals, a unique multivariate generalization of the new piecewise linear histogram is available. All these estimators are shown to be universally L1 strongly consistent. We derive large deviation inequalities. For twice differentiable densities with compact support their expected L1 error is shown to have the same rate of convergence as have kernel density estimators. Some simulated examples are presented.

AB - We study piecewise linear density estimators from the L1 point of view: the frequency polygons investigated by SCOTT (1985) and JONES et al. (1997), and a new piecewise linear histogram. In contrast to the earlier proposals, a unique multivariate generalization of the new piecewise linear histogram is available. All these estimators are shown to be universally L1 strongly consistent. We derive large deviation inequalities. For twice differentiable densities with compact support their expected L1 error is shown to have the same rate of convergence as have kernel density estimators. Some simulated examples are presented.

KW - Asymptotics

KW - Histogram

KW - Nonparametric density estimation

UR - http://www.scopus.com/inward/record.url?scp=0033238447&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033238447&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033238447

VL - 53

SP - 287

EP - 308

JO - Statistica Neerlandica

JF - Statistica Neerlandica

SN - 0039-0402

IS - 3

ER -