PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Using data compressors to construct order tests for homogeneity and component independence
Daniil Ryabko and Juergen Schmidhuber
Applied Mathematics Letters Volume 22, Number 7, pp. 1029-1032, 2009. ISSN 0893-9659

Abstract

Nonparametric order tests for homogeneity and component independence are proposed, which are based on data compressors. For homogeneity testing the idea is to compress the word obtained by ordering the combined samples and writing the number of the sample in the place of each element. H0 should be rejected if the string is compressed to a certain degree and accepted otherwise. We show that such a test obtained from an ideal data compressor is valid against all alternatives. Component independence is reduced to homogeneity testing.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
ID Code:5974
Deposited By:Daniil Ryabko
Deposited On:08 March 2010