Cause-effect inference through spectral independence in linear dynamical systems: theoretical foundations

TL;DR

This paper provides a theoretical foundation for the Spectral Independence Criterion (SIC) in cause-effect inference from time series, demonstrating its robustness and limitations through information theory, identifiability, and invariance analysis.

Contribution

It offers a rigorous theoretical justification for SIC, including identifiability results, robustness to downsampling, and insights into its limitations and potential generalizations.

Findings

SIC is supported by an information theoretic interpretation.

SIC remains effective under downsampling conditions.

Limitations of spectral independence are identified and avenues for generalization are discussed.

Abstract

Distinguishing between cause and effect using time series observational data is a major challenge in many scientific fields. A new perspective has been provided based on the principle of Independence of Causal Mechanisms (ICM), leading to the Spectral Independence Criterion (SIC), postulating that the power spectral density (PSD) of the cause time series is uncorrelated with the squared modulus of the frequency response of the filter generating the effect. Since SIC rests on methods and assumptions in stark contrast with most causal discovery methods for time series, it raises questions regarding what theoretical grounds justify its use. In this paper, we provide answers covering several key aspects. After providing an information theoretic interpretation of SIC, we present an identifiability result that sheds light on the context for which this approach is expected to perform well. We further demonstrate the robustness of SIC to downsampling - an obstacle that can spoil Granger-based inference. Finally, an invariance perspective allows to explore the limitations of the spectral independence assumption and how to generalize it. Overall, these results support the postulate of Spectral Independence is a well grounded leading principle for causal inference based on empirical time series.