Classical GP Algorithm for Correlation Dimension Computation

This repository provides source code implementing the classical Grassberger-Procaccia (GP) algorithm for calculating correlation dimensions - a fundamental method in nonlinear time series analysis. The implementation enables robust data analysis, particularly effective when processing large datasets. The GP algorithm operates by constructing correlation integrals to quantify the scaling properties of strange attractors in dynamical systems. Key implementation aspects include phase space reconstruction using time-delay embedding, distance matrix computation between state vectors, and logarithmic scaling analysis to extract the correlation dimension. Compared to alternative approaches, the classical GP algorithm offers advantages such as straightforward implementation, minimal parameter tuning requirements, and reliable convergence properties for various dynamical systems. The code structure includes modular functions for data preprocessing, correlation sum calculation, and dimension estimation through linear regression. For researchers and practitioners interested in nonlinear data analysis and chaos theory, this source code provides a valuable practical tool worth exploring.