A New Approach for Granger Causality between Neuronal Signals using the Empirical Mode Decomposition Algorithm

João Rodrigues and Alexandre Andrade

Keywords

Effective connectivity, Granger causality, inference, Empirical Mode Decomposition

Abstract

A new approach to measure causality between time series is presented as an alternative to the current time and frequency-domain Granger Causality. In this method the spectral information of causal relationships is not assessed by the spectrum of the coefficients of an autoregressive model and conditional time-domain Granger Causality can be used for its calculation. This is done by previously decomposing each data variable into its intrinsic oscillatory modes with the Empirical Mode Decomposition for posterior use of conditional time-domain causality. This work covers the most important theoretical background of these techniques and tries to prove the usefulness of this new approach while pointing out some of its qualities and possible drawbacks.

Important Links:

Go Back