CARIFORUM and UK EPA Study

regressions, there are a number of different approaches to testing for Granger Causality in a panel context. The different forms of panel causality test differ on the assumptions made about the homogeneity of the coefficients across cross-sections. EViews offers two of the simplest approaches to causality testing in panels. The first is to treat the panel data as one large stacked set of data, and then perform the Granger Causality test in the standard way, with the exception of not letting data from one cross-section enter the lagged values of data from the next cross-section. This method assumes that all coefficients are same across all cross-sections, i.e.: A second approach adopted by Dumitrescu-Hurlin (2012), makes an extreme opposite assumption, allowing all coefficients to be different across cross-sections: Here we show results for the pairwise Dumitrescu-Hurlin tests using data from “gasoline.WF1” (which is available in your examples directory). We reject the null that LCARPCAP does not homogeneously cause LGASPCAR (p=0.0015), but do not direct in the opposite direction. We subject the main variables of this study to a dual -causality test to evaluate the strength of our preliminary regression model of the volume of trade. We generally fail to reject the null, in some instances, that some of the variables do not Granger-cause the average trade volume of the region. We note, however, that the variables are collectively significant, with stronger preferences for the growth of the labor force and GDP (see the following table).

Page 214 of 241