Estimation of mutual information using kernel density estimators

Young Il Moon, Balaji Rajagopalan, Upmanu Lall

Research output: Contribution to journalArticlepeer-review

381 Scopus citations


Mutual information is useful for investigating the dependence between two experimental time series. It is often used to establish an appropriate time delay in phase-portrait reconstruction from time-series data. A histogram based approach has been used so far to estimate the probabilities. It is shown here that kernel density estimation of the probability density functions needed in estimating the average mutual information across two coordinates can be more effective than the histogram method of Fraser and Swinney [Phys. Rev. A 33, 1134 (1986)].

Original languageEnglish
Pages (from-to)2318-2321
Number of pages4
JournalPhysical Review E
Issue number3
StatePublished - 1995


Dive into the research topics of 'Estimation of mutual information using kernel density estimators'. Together they form a unique fingerprint.

Cite this