I am doing a research on the relation ship between FDI and GDP between 1980-2015 in one country. The values used are FDI stock and real gdp.

I ran the test using pcgive. But, for the optimal lag strucuture test, I get these results. Are these acceptable results? enter image description here

Also, using the optimal lag structure which is 2, I ran the Johansen co-integration test which says, I could not reject the hypothesis according to the p-value. The p-value of none is 0.1851 and p-value for at most 1 is 0.5069

Finally, I ran the granger causality test and get these results. I am quite confused here because according to what I learnt if we reject co-integration test, I should not have any rejection in the granger causality test also.

Real GDP does not Granger cause FDI Stock= 0.0011
FDI Stock does not Granger cause Real GDP= 0.0002

I also ran the ADF Test using first difference and it says I don't have any unit root. I don't know if these results would be acceptable or not.

Kindly advise me and h

  • $\begingroup$ What is your unit of analysis? Annual? Quarterly? And so on. $\endgroup$ – DJohnson Jan 30 '17 at 16:30
  • $\begingroup$ It's all annual data. $\endgroup$ – J.Eum Jan 30 '17 at 16:31
  • $\begingroup$ Absence of cointegration does not imply Granger noncausality. (Presence of cointegration does imply Granger causality at least one way, though.) $\endgroup$ – Richard Hardy Jan 30 '17 at 16:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.