Choosing the smoothing parameter in a fourier approach to nonparametric deconvolution of a density estimate

J. Barry, P. Diggle

Research output: Contribution to journalArticle

Abstract

In this note we derive a weighted non-linear least squares procedure for choosing the smoothing parameter in a Fourier approach to deconvolution of a density estimate. The method has the advantage over a previous procedure in that it is robust to the range of frequencies over which the model is fitted. A simulation study with different parametric forms for the densities in the convolution equation demonstrates that the method can perform well in practice. A truncated form of the estimator generally has a lower mean asymptotic integrated squared error than an alternative, continuously damped form, but the damped method gives better estimates of tail probabilities.

Original languageEnglish (US)
Pages (from-to)223-232
Number of pages10
JournalJournal of Nonparametric Statistics
Volume4
Issue number3
DOIs
StatePublished - Jan 1 1995
Externally publishedYes

Fingerprint

Density Estimates
Smoothing Parameter
Deconvolution
Damped
Integrated Squared Error
Convolution Equation
Nonlinear Least Squares
Tail Probability
Weighted Least Squares
Simulation Study
Estimator
Alternatives
Estimate
Range of data
Demonstrate
Form
Smoothing
Model

Keywords

  • Deconvolution
  • density estimation
  • Fourier transform
  • smoothing

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

Choosing the smoothing parameter in a fourier approach to nonparametric deconvolution of a density estimate. / Barry, J.; Diggle, P.

In: Journal of Nonparametric Statistics, Vol. 4, No. 3, 01.01.1995, p. 223-232.

Research output: Contribution to journalArticle

@article{395b6d1368cb4444a87def2c61be15d2,
title = "Choosing the smoothing parameter in a fourier approach to nonparametric deconvolution of a density estimate",
abstract = "In this note we derive a weighted non-linear least squares procedure for choosing the smoothing parameter in a Fourier approach to deconvolution of a density estimate. The method has the advantage over a previous procedure in that it is robust to the range of frequencies over which the model is fitted. A simulation study with different parametric forms for the densities in the convolution equation demonstrates that the method can perform well in practice. A truncated form of the estimator generally has a lower mean asymptotic integrated squared error than an alternative, continuously damped form, but the damped method gives better estimates of tail probabilities.",
keywords = "Deconvolution, density estimation, Fourier transform, smoothing",
author = "J. Barry and P. Diggle",
year = "1995",
month = "1",
day = "1",
doi = "10.1080/10485259508832614",
language = "English (US)",
volume = "4",
pages = "223--232",
journal = "Journal of Nonparametric Statistics",
issn = "1048-5252",
publisher = "Taylor and Francis Ltd.",
number = "3",

}

TY - JOUR

T1 - Choosing the smoothing parameter in a fourier approach to nonparametric deconvolution of a density estimate

AU - Barry, J.

AU - Diggle, P.

PY - 1995/1/1

Y1 - 1995/1/1

N2 - In this note we derive a weighted non-linear least squares procedure for choosing the smoothing parameter in a Fourier approach to deconvolution of a density estimate. The method has the advantage over a previous procedure in that it is robust to the range of frequencies over which the model is fitted. A simulation study with different parametric forms for the densities in the convolution equation demonstrates that the method can perform well in practice. A truncated form of the estimator generally has a lower mean asymptotic integrated squared error than an alternative, continuously damped form, but the damped method gives better estimates of tail probabilities.

AB - In this note we derive a weighted non-linear least squares procedure for choosing the smoothing parameter in a Fourier approach to deconvolution of a density estimate. The method has the advantage over a previous procedure in that it is robust to the range of frequencies over which the model is fitted. A simulation study with different parametric forms for the densities in the convolution equation demonstrates that the method can perform well in practice. A truncated form of the estimator generally has a lower mean asymptotic integrated squared error than an alternative, continuously damped form, but the damped method gives better estimates of tail probabilities.

KW - Deconvolution

KW - density estimation

KW - Fourier transform

KW - smoothing

UR - http://www.scopus.com/inward/record.url?scp=0041691699&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0041691699&partnerID=8YFLogxK

U2 - 10.1080/10485259508832614

DO - 10.1080/10485259508832614

M3 - Article

VL - 4

SP - 223

EP - 232

JO - Journal of Nonparametric Statistics

JF - Journal of Nonparametric Statistics

SN - 1048-5252

IS - 3

ER -