pdcEntropy: Permutation Distribution Entropy

In Jakob-Bach/FastTSDistances: Fast dissimilarity computations for time series

Description

More stable version of the entropy measure which is used by Brandmaier (2015) to determine the optimal embedding length in permutation distribution clustering. Mainly copies the (internal) function pdc::codebook.entropy, but also handles the case where only one pattern occurs in the time series.

Usage

pdcEntropy(timeSeries, subSequenceLength)

Arguments

timeSeries	A numeric vector/time series (minimum length: 2 elements).
subSequenceLength	Number of elements in subsequences which are used to count patterns.

Details

Encodes time series subsequences of the requested length, counts the relative frequency of each ordinal pattern and computes the entropy on this distribution. To obtain an unbiased estimate, Brandmaier proposes to divide by the log of non-zero-bin count.

Value

The normalized entropy as double.

References

Brandmaier, A. M. (2011). Permutation distribution clustering and structural equation model trees (Doctoral dissertation, Universität des Saarlandes, Saarbrücken).

Brandmaier, A. M. (2015). pdc: An R Package for Complexity-Based Clustering of Time Series. Journal of Statistical Software, 67(5), 1-23.

See Also

Other PDC functions: pdcDistTSListMult, pdcDistTSList, pdcDistTwoTSMult, pdcDistTwoTS, pdcEntropyHeuristicMult, pdcEntropyHeuristic

Jakob-Bach/FastTSDistances documentation built on May 13, 2019, 1:15 p.m.