Hi,

I just had a quick question about how to compare the results of different granger causality tests. I have been using the vargranger command to conduct granger causality tests after running vector autoregressions; however, I am not sure how to compare my results. I have several significant results, with some having lower p-values than others. Is there anything I can state about the tests with the lower (statistically significant) p-values? I.e., could I say that the variables with a lower p-value have more predictive power, or would this be untrue?

Thanks,
Amy