Two new Markov order estimators
Statistics Theory
2007-06-13 v1 Probability
Statistics Theory
Abstract
We present two new methods for estimating the order (memory depth) of a finite alphabet Markov chain from observation of a sample path. One method is based on entropy estimation via recurrence times of patterns, and the other relies on a comparison of empirical conditional probabilities. The key to both methods is a qualitative change that occurs when a parameter (a candidate for the order) passes the true order. We also present extensions to order estimation for Markov random fields.
Cite
@article{arxiv.math/0506080,
title = {Two new Markov order estimators},
author = {Yuval Peres and Paul Shields},
journal= {arXiv preprint arXiv:math/0506080},
year = {2007}
}
Comments
15 pages