Smoothed and iterated bootstrap confidence regions for parameter vectors

DOI10.1016/J.JMVA.2014.08.003zbMATH Open1360.62232arXiv1310.8339OpenAlexW2021853705MaRDI QIDQ458648FDOQ458648

Santu Ghosh, Alan M. Polansky

Publication date: 8 October 2014

Published in: Journal of Multivariate Analysis (Search for Journal in Brave)

Abstract: The construction of confidence regions for parameter vectors is a difficult problem in the nonparametric setting, particularly when the sample size is not large. The bootstrap has shown promise in solving this problem, but empirical evidence often indicates that some bootstrap methods have difficulty in maintaining the correct coverage probability, while other methods may be unstable, often resulting in very large confidence regions. One way to improve the performance of a bootstrap confidence region is to restrict the shape of the region in such a way that the error term of an expansion is as small an order as possible. To some extent, this can be achieved by using the bootstrap to construct an ellipsoidal confidence region. This paper studies the effect of using the smoothed and iterated bootstrap methods to construct an ellipsoidal confidence region for a parameter vector. The smoothed estimate is based on a multivariate kernel density estimator. This paper establishes a bandwidth matrix for the smoothed bootstrap procedure that reduces the asymptotic coverage error of the bootstrap percentile method ellipsoidal confidence region. We also provide an analytical adjustment to the nominal level to reduce the computational cost of the iterated bootstrap method. Simulations demonstrate that the methods can be successfully applied in practice.

Full work available at URL: https://arxiv.org/abs/1310.8339

zbMATH Keywords

Edgeworth expansionbandwidth matrixsmooth function modelbootstrap percentile methodbootstrap percentile-\(t\) methoditerated bootstrap method

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Nonparametric statistical resampling methods (62G09) Nonparametric tolerance and confidence regions (62G15)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Bootstrap methods: another look at the jackknife
• On the bootstrap and confidence intervals
• Theoretical comparison of bootstrap confidence intervals
• On the validity of the formal Edgeworth expansion
• Prepivoting to reduce level error of confidence sets
• Measures of multivariate skewness and kurtosis with applications
• Better Bootstrap Confidence Intervals
• Comparison of Smoothing Parameterizations in Bivariate Kernel Density Estimation
• Calibrating Confidence Coefficients
• The bootstrap and Edgeworth expansion
• Nonparametric standard errors and confidence intervals
• On bootstrap resampling and iteration
• Smoothed bootstrap confidence intervals with discrete data
• Bandwidth selection for the smoothed bootstrap percentile method.

Cited In (5)

• Iterated smoothed bootstrap confidence intervals for population quantiles
• The fundamental role of density functions in the binary classification problem
• Bootstrap confidence regions based on M-estimators under nonstandard conditions
• Bootstrapping analogs of the one way MANOVA test
• Large-scale simultaneous testing using kernel density estimation

Uses Software

• KernSmooth

This page was built for publication: Smoothed and iterated bootstrap confidence regions for parameter vectors

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q458648)