Asymptotics and optimal bandwidth selection for highest density region estimation

Abstract

We study kernel estimation of highest-density regions (HDR). Our main contributions are two-fold. First, we derive a uniform-in-bandwidth asymptotic approximation to a risk that is appropriate for HDR estimation. This approximation is then used to derive a bandwidth selection rule for HDR estimation possessing attractive asymptotic properties. We also present the results of numerical studies that illustrate the benefits of our theory and methodology.

Published in at http://dx.doi.org/10.1214/09-AOS766 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (2)

• Optimal rates for plug-in estimators of density level sets
• Asymptotic normality of plug-in level set estimates

Cited by in corpus (10)

• Optimal weighted nearest neighbour classifiers
• Clusters and water flows: a novel approach to modal clustering through Morse theory
• Adaptation to lowest density regions with application to support recovery
• Generalized Laplace Inference in Multiple Change-Points Models
• Change point analysis in non-stationary processes - a mass excess approach
• Asymptotics and Optimal Bandwidth for Nonparametric Estimation of Density Level Sets
• Minimax Hausdorff estimation of density level sets
• A fully data-driven method for estimating density level sets
• A comparative simulation study of data-driven methods for estimating density level sets
• Nonparametric Confidence Regions for Level Sets: Statistical Properties and Geometry