Marga8/HDGCvar: Granger Causality Testing in High Dimensional Vector Autoregressive Models

This package allows for testing Granger causality in High Dimensional Vector Autoregressive Models. The testing procedure is based on the Post Double Selection Lagrange Multiplier test developed in A. Hecq, L. Margaritella, S.Smeekes, "Granger Causality Testing in High Dimensional VARs: a Post Double Selection Procedure"(2019) and A. Hecq, L. Margaritella, S.Smeekes, "Inference in Non Stationary High Dimensional VARs" (2020). Granger causality can be tested between time series that are stationary, non stationary (unit root), cointegrated, or all the above. Bivariate as well as multivariate (i.e. blocks) causality can be considered and networks can be plotted. A specific part of the package is dedicated to Realized Volatility with the possibility of building spillover networks and condition on Realized Correlations.

install.packages("remotes")
remotes::install_github("Marga8/HDGCvar")