next up previous

KOLMOGOROV ENTROPY FROM TIME SERIES USING INFORMATION-THEORETIC FUNCTIONALS

Milan Palus
Institute of Computer Science, Academy of Sciences of the Czech Republic
Pod vodárenskou vezí 2, 182 07 Prague 8, Czech Republic
E-mail: mp@uivt.cas.cz, mp@santafe.edu

Abstract:

A technique for identification and quantification of chaotic dynamics in experimental time series is presented. It is based on evaluation of redundancies, information-theoretic functionals which, estimated from data generated by a chaotic dynamical system, have specific properties reflecting positive information production rate. This rate, measured by metric (Kolmogorov-Sinai) entropy, can be directly estimated from the redundancies.



Neural Network World Vol. 7 No. 3 (1997) 269-292



Milan Palus
January-August 1997