A measure of statistical complexity based on predictive information
Statistics Theory
2010-12-10 v1 Information Theory
math.IT
Data Analysis, Statistics and Probability
Statistics Theory
Abstract
We introduce an information theoretic measure of statistical structure, called 'binding information', for sets of random variables, and compare it with several previously proposed measures including excess entropy, Bialek et al.'s predictive information, and the multi-information. We derive some of the properties of the binding information, particularly in relation to the multi-information, and show that, for finite sets of binary random variables, the processes which maximises binding information are the 'parity' processes. Finally we discuss some of the implications this has for the use of the binding information as a measure of complexity.
Cite
@article{arxiv.1012.1890,
title = {A measure of statistical complexity based on predictive information},
author = {Samer A. Abdallah and Mark D. Plumbley},
journal= {arXiv preprint arXiv:1012.1890},
year = {2010}
}
Comments
4 pages, 3 figures