Uniform in bandwidth exact rates for a class of kernel estimators

DOI10.1007/S10463-010-0286-5MaRDI QIDQ652606zbMATH OpenOpenAlexFDO

Authors Davit Varron, Ingrid Van Keilegom

Publication date 14 December 2011

Published in Annals of the Institute of Statistical Mathematics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1201.5507

zbMATH Keywords

empirical likelihoodkernel smoothinglocal empirical processesuniform in bandwidth consistency

Mathematics Subject Classification ID

Strong limit theorems (60F15) Nonparametric estimation (62G05) Density estimation (62G07) Nonparametric tolerance and confidence regions (62G15) Asymptotic properties of nonparametric inference (62G20)

Abstract: Given an i.i.d sample $(Y_i,Z_i)$, taking values in $RR{R}^{d^{\prime }}imesRR{R}^d$, we consider a collection Nadarya-Watson kernel estimators of the conditional expectations $EEE(<c_g(z),g(Y)>+d_g(z)midZ=z)$, where $z$ belongs to a compact set $HsubsetRR{R}^d$, $g$ a Borel function on $RR{R}^{d^{\prime }}$ and $c_g(cdot),d_g(cdot)$ are continuous functions on $RR{R}^d$. Given two bandwidth sequences $h_n<wt{h}_n$ fulfilling mild conditions, we obtain an exact and explicit almost sure limit bounds for the deviations of these estimators around their expectations, uniformly in $ginGG,;zinH$ and $h_{n}lehlewt{h}_n$ under mild conditions on the density $f_Z$, the class $GG$, the kernel $K$ and the functions $c_g(cdot),d_g(cdot)$. We apply this result to prove that smoothed empirical likelihood can be used to build confidence intervals for conditional probabilities $PPP(YinCmidZ=z)$, that hold uniformly in $zinH,;CinCC,;hin[h_n,wt{h}_n]$. Here $CC$ is a Vapnik-Chervonenkis class of sets.

Recommendations

• Uniform in bandwidth consistency of kernel-type function estimators
• Uniform central limit theorems for kernel density estimators
• scientific article; zbMATH DE number 3947418
• Uniform in bandwidth consistency of conditional \(U\)-statistics
• General asymptotic confidence bands based on kernel-type function estimators

Cites work

• A Bennett concentration inequality and its application to suprema of empirical processes
• A limited in bandwidth uniformity for the functional limit law of the increments of the empirical process
• A uniform functional law of the logarithm for the local empirical process.
• An Empirical Likelihood Goodness-of-Fit Test for Time Series
• An empirical process approach to the uniform consistency of kernel-type function estimators
• Empirical likelihood
• How Far Are Automatically Chosen Regression Smoothing Parameters From Their Optimum?
• scientific article; zbMATH DE number 3415224 (Why is no real title available?)
• On the Choice of the Bandwidth in Kernel Nonparametric Regression
• Some universal results on the behavior of increments of partial sums
• Strong approximation for multivariate empirical and related processes, via KMT constructions
• Uniform in bandwidth consistency of kernel-type function estimators
• Weak convergence and empirical processes. With applications to statistics

Cited in

(4)

• Functional limit laws for local empirical processes in a spatial setting
• Uniform-in-bandwidth functional limit laws
• Uniform in bandwidth consistency for various kernel estimators involving functional data
• A Donsker and Glivenko-Cantelli theorem for random measures linked to extreme value theory

This page was built for publication: Uniform in bandwidth exact rates for a class of kernel estimators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q652606)