Skip to main content
SpringerLink
Log in

Exact and Asymptotically Optimal Bandwidths for Kernel Estimation of Density Functionals

  • Published:
Methodology and Computing in Applied Probability Aims and scope Submit manuscript

Abstract

Given a density f we pose the problem of estimating the density functional \(\psi_r=\int f^{(r)}f\) for a non-negative even r making use of kernel methods. This is a well-known problem but some of its features remained unexplored. We focus on the problem of bandwidth selection. Whereas all the previous studies concentrate on an asymptotically optimal bandwidth here we study the properties of exact, non-asymptotic ones, and relate them with the former. Our main conclusion is that, despite being asymptotically equivalent, for realistic sample sizes much is lost by using the asymptotically optimal bandwidth. In contrast, as a target for data-driven selectors we propose another bandwidth which retains the small sample performance of the exact one.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Aldershof B (1991) Estimation of integrated squared density derivatives. PhD thesis, University of North Carolina, Chapel Hill

  • Aldershof B, Marron JS, Park BU, Wand MP (1995) Facts about the Gaussian probability density function. Appl Anal 59:289–306

    Article  MathSciNet  MATH  Google Scholar 

  • Bhattacharya GK, Roussas GG (1969) Estimation of a certain functional of a probability density function. Skand Aktuarietidskr 1969:201–206

    MathSciNet  Google Scholar 

  • Bickel PJ, Ritov Y (1988) Estimating integrated squared density derivatives: sharp best order of convergence estimates. Sankhya Ser 50:381–393

    MathSciNet  MATH  Google Scholar 

  • Butzer PL, Nessel RJ (1971) Fourier analysis and approximation, vol 1. Birkhäuser, Basel

    Book  MATH  Google Scholar 

  • Chacón JE, Tenreiro C (2011) Data-based choice of the number of pilot stages for plug-in bandwidth selection. To appear in Commun Statist Theory Methods. doi:10.1080/03610926.2011.606486

    Google Scholar 

  • Chacón JE, Montanero J, Nogales AG (2007a) A note on kernel density estimation at a parametric rate. J Nonparametr Stat 19:13–21

    Article  MathSciNet  MATH  Google Scholar 

  • Chacón JE, Montanero J, Nogales AG, Pérez P (2007b) On the existence and limit behavior of the optimal bandwidth for kernel density estimation. Stat Sin 17:289–300

    MATH  Google Scholar 

  • Dmitriev YG, Tarasenko FP (1973) On the estimation of functionals of the probability density and its derivatives. Theory Probab Appl 18:628–633

    Article  MathSciNet  MATH  Google Scholar 

  • Dmitriev YG, Tarasenko FP (1975) On a class of non-parametric estimates of non-linear functionals of density. Theory Probab Appl 19:390–394

    Article  Google Scholar 

  • Giné E, Mason DM (2008) Uniform in bandwidth estimation of integral functionals of the density function. Scand J Statist 35:739–761

    Article  MathSciNet  MATH  Google Scholar 

  • Giné E, Nickl R (2008) A simple adaptive estimator of the integrated square of a density. Bernoulli 14:47–61

    Article  MathSciNet  MATH  Google Scholar 

  • Hall P, Marron JS (1987) Estimation of integrated squared density derivatives. Stat Probab Lett 6:109–115

    Article  MathSciNet  MATH  Google Scholar 

  • Hettmansperger TP (1984) Statistical inference based on ranks. Wiley, New York

    MATH  Google Scholar 

  • Hewitt E, Stromberg K (1965) Real and abstract analysis. Springer, Berlin

    Book  MATH  Google Scholar 

  • Jones MC, Sheather SJ (1991) Using non-stochastic terms to advantage in kernel-based estimation of integrated squared density derivatives. Stat Probab Lett 11:511–514

    Article  MathSciNet  MATH  Google Scholar 

  • Laurent B (1997) Estimation of integral functionals of a density and its derivatives. Bernoulli 3:181–211

    Article  MathSciNet  MATH  Google Scholar 

  • Levit BY (1978) Asymptotically efficient estimation of nonlinear functionals. Probl Inf Transm 14:65–72

    MathSciNet  Google Scholar 

  • Marron JS, Wand MP (1992) Exact mean integrated squared error. Ann Stat 20:712–736

    Article  MathSciNet  MATH  Google Scholar 

  • Politis DN, Romano JP (1999) Multivariate density estimation with general flat-top kernels of infinite order. J Multivar Anal 68:1–25

    Article  MathSciNet  MATH  Google Scholar 

  • Prakasa Rao BLS (1983) Nonparametric functional estimation. Academic, New York

    MATH  Google Scholar 

  • Prakasa Rao BLS (1999) Estimation of the integrated squared density derivatives by wavelets. Bull Inform Cybern 31:47–65

    MathSciNet  MATH  Google Scholar 

  • Scott DW (1992) Multivariate density estimation: theory, practice and visualization. Wiley, New York

    Book  MATH  Google Scholar 

  • Schuster EF (1974) On the rate of convergence of an estimate of a functional of a probability density. Scand Actuar J 1974:103–107

    Article  MathSciNet  MATH  Google Scholar 

  • Sheather SJ, Hettmansperger TP, Donald MR (1994) Data-based bandwidth selection for kernel estimators of the integral of f 2(x). Scand J Statist 21:265–275

    MathSciNet  MATH  Google Scholar 

  • Tenreiro, C (2003) On the asymptotic normality of multistage integrated density derivatives kernel estimators. Stat Probab Lett 64:311–322

    Article  MathSciNet  MATH  Google Scholar 

  • van Es B (1992) Estimating functionals related to a density by a class of statistics based on spacings. Scand J Statist 19:61–72

    MathSciNet  MATH  Google Scholar 

  • Wand MP, Jones MC (1995) Kernel smoothing. Chapman and Hall, London

    MATH  Google Scholar 

  • Wand MP, Schucany WR (1990) Gaussian-based kernels. Can J Stat 18:197–204

    Article  MathSciNet  Google Scholar 

  • Wu T-J (1995) Adaptive root n estimates of integrated squared density derivatives. Ann Stat 23:1474–1495

    Article  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Departamento de Matemáticas, Universidad de Extremadura, Badajoz, Spain

    José E. Chacón

  2. CMUC, Department of Mathematics, University of Coimbra, Coimbra, Portugal

    Carlos Tenreiro

Authors
  1. José E. Chacón
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Carlos Tenreiro
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to José E. Chacón.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Chacón, J.E., Tenreiro, C. Exact and Asymptotically Optimal Bandwidths for Kernel Estimation of Density Functionals. Methodol Comput Appl Probab 14, 523–548 (2012). https://doi.org/10.1007/s11009-011-9243-x

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11009-011-9243-x

Keywords

AMS 2000 Subject Classifications

Search

Navigation