On the consistency of the Kozachenko-Leonenko entropy estimate

Abstract
We revisit the problem of the estimation of the differential entropy of a random vector in with density , assuming that exists and is finite. In this note, we study the consistency of the popular nearest neighbor estimate of Kozachenko and Leonenko. Without any smoothness condition we show that the estimate is consistent ( as ) if and only if . Furthermore, if has compact support, then almost surely.
View on arXivComments on this paper