Graphs in State-Space Models for Granger Causality in Climate Science
This work addresses the need for improved causality analysis in climate science, though it appears incremental as it builds on existing GraphEM and regularization techniques.
The authors tackled the problem of assessing Granger causality in time series by proposing a graphical state-space model approach using GraphEM with Lasso regularization, which demonstrated benefits over standard methods in toy examples and climate applications.
Granger causality (GC) is often considered not an actual form of causality. Still, it is arguably the most widely used method to assess the predictability of a time series from another one. Granger causality has been widely used in many applied disciplines, from neuroscience and econometrics to Earth sciences. We revisit GC under a graphical perspective of state-space models. For that, we use GraphEM, a recently presented expectation-maximisation algorithm for estimating the linear matrix operator in the state equation of a linear-Gaussian state-space model. Lasso regularisation is included in the M-step, which is solved using a proximal splitting Douglas-Rachford algorithm. Experiments in toy examples and challenging climate problems illustrate the benefits of the proposed model and inference technique over standard Granger causality methods.