Maximum likelihood estimation of a multidimensional log-concave density

Let X_1, ..., X_n be independent and identically distributed random vectors with a log-concave (Lebesgue) density f. We first prove that, with probability one, there exists a unique maximum likelihood estimator of f. The use of this estimator is attractive because, unlike kernel density estimation,...

Fuld beskrivelse

Bibliografiske detaljer
Main Authors: Cule, M, Samworth, R, Stewart, M
Format: Journal article
Sprog:English
Udgivet: Blackwell Publishing Ltd 2008

Lignende værker