I'm getting different answers everywhere I look, including this site.

I'm using granger causality to substantiate cointegration and therefore the use of a VECM model.

Specifically I am using the grangercausalitytests from statsmodels.

Do I perform the test on the original data (i.e., GDP is 1.5 BN then 2.5 BN...) or the stationary data (i.e., the differenced data assuming I(1))?

Thank you

  • 1
    $\begingroup$ See Dave Giles' blog post for an excellent overview. $\endgroup$ – Richard Hardy Apr 15 at 5:53

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.