Testing for Granger Causality in Panel Data

With the development of large and long panel databases, the theory surrounding panel causality evolves at a fast pace and empirical researchers may sometimes find it difficult to run the most recent techniques developed in the literature. This article presents the Stata user-written command xtgcause, which implements a procedure proposed by Dumitrescu & Hurlin (2012) for detecting Granger causality in panel datasets, and thus constitutes an effort to help practitioners understand and apply the test. The command offers the possibility to select the number of lags to include in the model by minimizing the AIC, BIC, or HQIC, and to implement a bootstrap procedure to compute p-values and critical values.