FUNDAMENTALNAYA I PRIKLADNAYA MATEMATIKA

(FUNDAMENTAL AND APPLIED MATHEMATICS)

2013, VOLUME 18, NUMBER 2, PAGES 209-227

Selection of a metric for the nearest neighbor entropy estimators

E. A. Timofeev

Abstract

View as HTML     View as gif image

We consider the problem of improving the efficiency of the nonparametric entropy estimation for a stationary ergodic process. Our approach is based on the nearest-neighbor distances. We propose a broad class of metrics on the space of right-sided infinite sequences drawn from a finite alphabet. The new metric has a parameter which is a nonincreasing function. We prove that, under certain conditions, our estimators have a small variance and show that a special selection of the metric parameters reduces the estimator's bias.

Location: http://mech.math.msu.su/~fpm/eng/k13/k132/k13217h.htm
Last modified: January 7, 2014