ordpy

ordpy implements permutation entropy and ordinal network methods to quantify complexity and analyze dynamics in time series and two-dimensional data.

Key Features:

• Permutation Entropy: Computes permutation entropy by mapping time series into ordinal patterns to quantify dynamic complexity.
• Tsallis and Rényi Permutation Entropies: Implements Tsallis and Rényi generalizations of permutation entropy to provide alternative complexity measures.
• Complexity-Entropy Plane and Curves: Generates complexity–entropy planes and curves for visualization of the relationship between entropy and complexity.
• Missing Ordinal Patterns and Transitions: Detects missing ordinal patterns and ordinal transitions to analyze incomplete or irregular sequences.
• Ordinal Networks: Constructs ordinal networks from time series to study structural properties and dynamics using network representations.
• Multiscale Generalizations: Supports multiscale analysis to compute ordinal measures across multiple temporal or spatial scales.
• Two-Dimensional Data Analysis: Applies permutation entropy and ordinal methods to two-dimensional data such as images.

Scientific Applications:

• Physics: Characterizes dynamical regimes and complexity in physical systems using permutation entropy and ordinal networks.
• Biology: Analyzes biological time series and image data to assess complexity and dynamical patterns.
• Finance: Quantifies complexity and detects dynamical changes in financial time series.
• Engineering: Evaluates complex system dynamics and multiscale phenomena in engineering datasets.

Methodology:

Implements algorithms based on Bandt and Pompe's permutation entropy framework, extensions to Tsallis and Rényi entropies, ordinal network construction, detection of missing ordinal patterns and transitions, and multiscale generalizations, with support for two-dimensional data.