Estimation of density level sets with a given probability content

@inproceedings{Cadrea2014EstimationOD,
title={Estimation of density level sets with a given probability content},
author={Beno{\^i}t Cadrea and Bruno Pelletierb and Pierre Pudloc},
year={2014}
}

Benoît Cadrea, Bruno Pelletierb, Pierre Pudloc

Published 2014

Given a random vector X valued in Rd with density f and an arbitrary probability number p ∈ (0; 1), we consider the estimation of the upper level set {f ≥ t(p)} of f corresponding to probability content p, that is, such that the probability that X belongs to {f ≥ t(p)} is equal to p. Based on an i.i.d. random sample X1, . . . , Xn drawn from f , we define the plug-in level set estimate {f̂n ≥ t n }, where t n is a random threshold depending on the sample and f̂n is a nonparametric kernel… CONTINUE READING

A Probabilistic Theory of Pattern Recognition, New-York: Springer-Verlag

L. Devroye, L. Gyorfi, G. Lugosi

Australian & New Zealand Journal of Statistics, • 1996

View 4 Excerpts

Highly Influenced

Computing and Graphing Highest Density Regions

R. Hyndman

The American Statistician, • 1996

View 3 Excerpts

Highly Influenced

Now, let J be any interval included in I . Observe that f is a locally Lipschitz function and that 1{f ∈ J} is integrable. According to Theorem 2, Chapter 3 in Evans and Gariepy