On optimal kernel choice for deconvolution

Aurore Delaigle, Peter Hall*

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    33 Citations (Scopus)

    Abstract

    In this note we show that, from a conventional viewpoint, there are particularly close parallels between optimal-kernel-choice problems in non-parametric deconvolution, and their better-understood counterparts in density estimation and regression. However, other aspects of these problems are distinctly different, and this property leads us to conclude that "optimal" kernels do not give satisfactory performance when applied to deconvolution. This unexpected result stems from the fact that standard side conditions, which are used to ensure that the familiar kernel-choice problem has a unique solution, do not have statistically beneficial implications for deconvolution estimators. In consequence, certain "sub-optimal" kernels produce estimators that enjoy both greater efficiency and greater visual smoothness.

    Original languageEnglish
    Pages (from-to)1594-1602
    Number of pages9
    JournalStatistics and Probability Letters
    Volume76
    Issue number15
    DOIs
    Publication statusPublished - 1 Sept 2006

    Fingerprint

    Dive into the research topics of 'On optimal kernel choice for deconvolution'. Together they form a unique fingerprint.

    Cite this