Welcome to Statistical Complexity Measures!

statcomp: An R package to quantify statistical complexity and information of time series to distinguish chaos from noise.

Contents

Introduction

Statistical complexity and information measures are statistical tools to quantify entropy and complexity in time series and hence to distinguish deterministic chaos from randomness (see e.g. Bandt and Pompe 2002). The measures are based on the “ordinal pattern distribution” of the time series (an alternative to a histogram-like representation with some advantages). We specifically provide a range of different permutation coding schemes for calculating the Fisher Information (see e.g. Olivares et al 2012). In addition, measures to quantify the statistical distance between ordinal pattern distributions are provided (e.g. Hellinger Distance).

Related papers:

Lopez-Ruiz, R., Mancini, H. L. & Calbet, X., 1995. A statistical measure of complexity. Physics Letters A, 209. doi:10.1016/0375-9601(95)00867-5.
Bandt, C. and Pompe, B., 2002. Permutation entropy: a natural complexity measure for time series. Physical Review Letters, 88(17). doi:10.1103/PhysRevLett.88.174102.
Rosso, O. A., Larrondo, H. A., Martin, M. T., Plastino, A., & Fuentes, M. A. (2007). Distinguishing noise from chaos. Physical Review Letters, 99(15). doi:10.1103/PhysRevLett.99.154102.
Olivares, F., Plastino, A. and Rosso, O.A., 2012. Ambiguities in Bandt-Pompe's methodology for local entropic quantifiers. Physica A: Statistical Mechanics and its Applications, 391(8). doi:10.1016/j.physa.2011.12.033.
Sippel, S., Lange, H., Mahecha, M. D., Hauhs, M., Bodesheim, P., Kaminski, T., Gans, F. & Rosso, O. A. (2016) Diagnosing the Dynamics of Observed and Simulated Ecosystem Gross Primary Productivity with Time Causal Information Theory Quantifiers. PLOS ONE, accepted. doi:10.1371/journal.pone.0164960.

Figure 1. Exemplary illustration of different time series in the entropy-complexity plane.
Figure 1. Exemplary illustration of different time series in the entropy-complexity plane.

Installation and Usage

To install statcomp, please type from your preferred R-console:
install.packages("statcomp", repos="http://R-Forge.R-project.org")

… or retrieve from CRAN (https://cran.r-project.org/web/packages/statcomp/ ):
install.packages("statcomp")

If you use statcomp in scientific publications, please cite:
Sippel, S., Lange, H., Mahecha, M. D., Hauhs, M., Bodesheim, P., Kaminski, T., Gans, F. & Rosso, O. A. (2016) Diagnosing the Dynamics of Observed and Simulated Ecosystem Gross Primary Productivity with Time Causal Information Theory Quantifiers. PLOS ONE, accepted. doi:10.1371/journal.pone.0164960.

The package has been developed at the Max Planck Institute for Biogeochemistry, Jena, Germany.

Author details and further information:

Sebastian Sippel (ssippel@bgc-jena.mpg.de)
For news and further information check my personal page: here.