Permutation entropy
The permutation entropy is a complexity parameter that was introduced by Bandt and Pompe in 2002 for observable measurements of chaotic dynamic systems, i.e. time series . It can be classified in the ordinal time series analysis, since it is based on the comparison of neighboring values. Thus the permutation entropy is equal to the Shannon entropy for ordinal patterns. The definition applies directly to any real data. The advantages of the method are its simplicity, extremely fast calculation, robustness and invariance with regard to nonlinear monotonic transformations.
literature
- Bandt, Christoph & Pompe, Bernd. (2002). Permutation Entropy: A Natural Complexity Measure for Time Series. In: Physical Review Letters. 88. 174102. doi : 10.1103 / PhysRevLett.88.174102 .
Individual evidence
- ↑ Alexander Benedikt Thul: Permutation entropy and symbolic transfer entropy as EEG - parameters for differentiating Disorders of Consciousness under auditory stimulation . Ed .: TUM - Technical University of Munich. 2017, p. 38–39 , urn : nbn: de: bvb: 91-diss-20171116-1275422-1-9 .