The Annals of Statistics

Formulae for Mean Integrated Squared Error of Nonlinear Wavelet-Based Density Estimators

Peter Hall and Prakash Patil

Full-text: Open access

Abstract

We provide an asymptotic formula for the mean integrated squared error (MISE) of nonlinear wavelet-based density estimators. We show that, unlike the analogous situation for kernel density estimators, this MISE formula is relatively unaffected by assumptions of continuity. In particular, it is available for densities which are smooth in only a piecewise sense. Another difference is that in the wavelet case the classical MISE formula is valid only for sufficiently small values of the bandwidth. For larger bandwidths MISE assumes a very different form and hardly varies at all with changing bandwidth. This remarkable property guarantees a high level of robustness against oversmoothing, not encountered in the context of kernel methods. We also use the MISE formula to describe an asymptotically optimal empirical bandwidth selection rule.

Article information

Source
Ann. Statist., Volume 23, Number 3 (1995), 905-928.

Dates
First available in Project Euclid: 11 April 2007

Permanent link to this document
https://projecteuclid.org/euclid.aos/1176324628

Digital Object Identifier
doi:10.1214/aos/1176324628

Mathematical Reviews number (MathSciNet)
MR1345206

Zentralblatt MATH identifier
0839.62042

JSTOR
links.jstor.org

Subjects
Primary: 62G07: Density estimation
Secondary: 62G20: Asymptotic properties

Keywords
Bandwidth density estimation kernel mean integrated squared error multiresolution robustness wavelet

Citation

Hall, Peter; Patil, Prakash. Formulae for Mean Integrated Squared Error of Nonlinear Wavelet-Based Density Estimators. Ann. Statist. 23 (1995), no. 3, 905--928. doi:10.1214/aos/1176324628. https://projecteuclid.org/euclid.aos/1176324628


Export citation