A general theory to estimate Information transfer in nonlinear systems

Abstract

A general theory for computing information transfers in nonlinear stochastic systems driven by deterministic forcing and additive and/or multiplicative noises, is presented. It extends the Liang-Kleeman framework of causality inference to nonlinear cases based on information transfer across system variables (Liang, 2016). We present an effective method of computing formulas of the rates of Shannon entropy transfer (RETs) between selected causal and consequential variables, relying on the estimation from data of conditional expectations of the system forcing and their derivatives. Those expectations are approximated by nonlinear differentiable regressions, leading to a much easier and more robust way of computing RETs than the brute-force approach calling for numerical integrals over the state-space and the knowledge of the multivariate probability density of the system. The approach is fully adapted to the case where no model equations are available, starting with a nonlinear model fitting from data of the consequential variables, and the subsequent application of method to the fitted model. RETs are decomposed into sums of single one-to-one RETs plus synergetic terms (of pure nonlinear nature) accounting for the joint causal effect of groups of variables. State-dependent RET formulas are also introduced, showing where in state-space the entropy transfers and local synergies are more relevant. A comparison of the RETs estimations is performed with previous methods in the context of two models: (i) a model derived from a potential function and (ii) the classical chaotic Lorenz system, both forced by additive and/or multiplicative noises. The analysis demonstrates that the new estimations are robust, cheaper, and less data-demanding, providing evidence of the possibilities and generalizations offered by the method and opening new perspectives on real-world applications.