Effective EEG Connectivity by Sparse Vector Autoregressive Model

Abhishek Goyal, Rahul Garg

Abstract: This paper introduces a time domain approach based on Granger causality for estimating directional flow between multivariate time series. It is formulated under the framework of vector autoregressive model. Sparse regression is used to find the solution to the VAR model and validation of the results are carried out with the help of simulations. We also demonstrate the application of this method on actual EEG dataset.