Rtractor: Complexity and Nonlinear Time Series Analysis for Physiological Signals

A staging- and signal-agnostic toolkit of complex systems measures for physiological time series: entropy (sample, permutation, approximate), fractal and multifractal analysis (Higuchi dimension, DFA, MFDFA), Lyapunov exponents, multiscale entropy, and recurrence quantification analysis. Core algorithms wrap established C/C++/Fortran reference implementations rather than reimplementing them in pure R, to preserve numerical parity with the original methods literature. Some wrapped reference implementations are GPL-licensed (see inst/COPYRIGHTS), which is why the package as a whole is licensed GPL (>= 2) rather than MIT. Designed to sit underneath other Circadia Lab and CoDe-Neuro Lab packages (mrpheus, hypnoR, zeitR, dynR) as a shared complexity-metrics layer, while remaining fully usable standalone on any numeric time series.

See also

Useful links:

• https://rtractor.circadia-lab.uk

• https://github.com/circadia-bio/Rtractor

• Report bugs at https://github.com/circadia-bio/Rtractor/issues

Author

Maintainer: Lucas França lucas.franca@northumbria.ac.uk (ORCID)

Authors:

• Lucas França lucas.franca@northumbria.ac.uk (ORCID)

• Mario Leocadio-Miguel mario.miguel@northumbria.ac.uk (ORCID)