irreversibility: A Python Package for Assessing and Manipulating the Time Irreversibility of Real-World Time Series

Time irreversibility refers to the property of some dynamical systems and time series of being statistically different when observed backward in time. While the theoretical foundations of irreversibility date back to the origin of statistical physics, the analysis of such property in real-world time...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Entropy (Basel, Switzerland) Jg. 27; H. 11; S. 1146
1. Verfasser: Zanin, Massimiliano
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Switzerland MDPI AG 12.11.2025
Schlagworte:
Asymmetry
Causality
Python package
Statistical physics
statistical test
Symmetry
time irreversibility
Time series
time series
statistical test
Python package
time irreversibility
ISSN:1099-4300, 1099-4300
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Time irreversibility refers to the property of some dynamical systems and time series of being statistically different when observed backward in time. While the theoretical foundations of irreversibility date back to the origin of statistical physics, the analysis of such property in real-world time series has only recently gained momentum. We present irreversibility, an open-source Python (version ≥ 3.11) package aimed at providing a large set of irreversibility metrics and tests and at facilitating their use. Besides the tests themselves, it includes a set of utilities, like functions to downsample and manipulate the time series, and to optimise the parameters of the metrics. By providing a unified software package, irreversibility simplifies the analysis of real data, allowing the researcher to compare multiple tests and obtain a better and more reproducible view of the underlying system. In this contribution we explore the features of the package and provide examples of its use.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1099-4300
1099-4300
DOI:10.3390/e27111146