Open Access
August 2008 Statistics of extremes by oracle estimation
Ion Grama, Vladimir Spokoiny
Ann. Statist. 36(4): 1619-1648 (August 2008). DOI: 10.1214/07-AOS535

Abstract

We use the fitted Pareto law to construct an accompanying approximation of the excess distribution function. A selection rule of the location of the excess distribution function is proposed based on a stagewise lack-of-fit testing procedure. Our main result is an oracle type inequality for the Kullback–Leibler loss.

Citation

Download Citation

Ion Grama. Vladimir Spokoiny. "Statistics of extremes by oracle estimation." Ann. Statist. 36 (4) 1619 - 1648, August 2008. https://doi.org/10.1214/07-AOS535

Information

Published: August 2008
First available in Project Euclid: 16 July 2008

zbMATH: 1282.62131
MathSciNet: MR2435450
Digital Object Identifier: 10.1214/07-AOS535

Subjects:
Primary: 62G08 , 62G32
Secondary: 62G05

Keywords: Extreme values , high quantiles , Hill estimator , Nonparametric adaptive estimation , probabilities of rare events

Rights: Copyright © 2008 Institute of Mathematical Statistics

Vol.36 • No. 4 • August 2008
Back to Top