Paper #505

Title:

A note on robust detection

Authors:

Luc Devroye, László Györfi and Gábor Lugosi

Date:

August 2000

Abstract:

We introduce a simple new hypothesis testing procedure, which, based on an independent sample drawn from a certain density, detects which of $k$ nominal densities is the true density is closest to, under the total variation (L_{1}) distance. We obtain a density-free uniform exponential bound for the probability of false detection.

Keywords:

Robust detection, hypotheses testing

JEL codes:

C13, C14

Area of Research:

Statistics, Econometrics and Quantitative Methods

Published in:

IEEE Transactions on Information Theory, vol. 48, pp.2111--2114, 2002

Download the paper in PDF format