Long-term Granger Causality Testing

[NAB01]

Hi,

I am preparing a time-series analysis using E-Views 7 and have a query about long-term granger causality testing, and would greatly appreciate your feedback.

I have already tested the long-run cointegration amongst the variables under examination using the Johansen technique and it resulted that the variables are in fact cointegrated, from which the ECT terms for each cointegrating relationship resulted (ie the cointegrating equation -Beta and the speed of adjustment - Alpha).

Moving on to Granger causality testing, I am aware of the function in Eviews whereby you select the variables under examination and perform the standard Granger Causality Tests for specific lag lengths. This test tests the Granger Causality between two variables (for example X and Y), by taking into account the lagged terms of X and the lagged terms of Y.

However, from my knowledge, if the variables are found to be cointegrated, in order to test the Granger Causality in the long-term, the ECT term/s also needs to be taken into account (in addition to the lagged terms of X and Y). Can you kindly guide me how this test may be performed using EViews 7?

Whilst awaiting for your feedback, I would like to thank you in advance for your assistance.

Kind regards,
Nadine