Partial Information Rate Decomposition

• DOI: 10.48550/arxiv.2502.04550
• Publication Date: 2025-02-06
• Authors: Luca Faes; Laura Sparacino; Gorana Mijatović; Yuri Antonacci; Leonardo Ricci; Daniele Marinazzo; Sebastiano Strama...
• Keywords: Methodology (stat.ME); FOS: Computer and information sciences; Statistics - Methodology

Partial Information Decomposition (PID) is a principled and flexible method to unveil complex high-order interactions in multi-unit network systems. Though being defined exclusively for random variables, PID ubiquitously applied multivariate time series taken as realizations of processes with temporal statistical structure. Here, overcome the incorrect depiction effects by schemes dynamic networks, we introduce framework Rate (PIRD). PIRD formalized applying lattice theory decompose information shared dynamically between target process set source processes, implemented Gaussian through spectral expansion rates, demonstrated practice analyzing from large-scale climate oscillations.