Current browse context:
math.ST
Change to browse by:
References & Citations
Mathematics > Statistics Theory
Title: Nearest neighbor density functional estimation from inverse Laplace transform
(Submitted on 22 May 2018 (v1), last revised 5 Feb 2022 (this version, v4))
Abstract: A new approach to $L_2$-consistent estimation of a general density functional using $k$-nearest neighbor distances is proposed, where the functional under consideration is in the form of the expectation of some function $f$ of the densities at each point. The estimator is designed to be asymptotically unbiased, using the convergence of the normalized volume of a $k$-nearest neighbor ball to a Gamma distribution in the large-sample limit, and naturally involves the inverse Laplace transform of a scaled version of the function $f.$ Some instantiations of the proposed estimator recover existing $k$-nearest neighbor based estimators of Shannon and R\'enyi entropies and Kullback--Leibler and R\'enyi divergences, and discover new consistent estimators for many other functionals such as logarithmic entropies and divergences. The $L_2$-consistency of the proposed estimator is established for a broad class of densities for general functionals, and the convergence rate in mean squared error is established as a function of the sample size for smooth, bounded densities.
Submission history
From: Jongha Jon Ryu [view email][v1] Tue, 22 May 2018 01:19:30 GMT (109kb)
[v2] Sat, 1 Aug 2020 06:18:22 GMT (9489kb,D)
[v3] Fri, 14 Aug 2020 00:05:06 GMT (9489kb,D)
[v4] Sat, 5 Feb 2022 01:54:27 GMT (10547kb,D)
Link back to: arXiv, form interface, contact.