My question is twofold (hope it's ok). I want to estimate VAR model with the sole purpose of analysing the impulse response functions. I want to analyse the response of GDP to shock in exports and consumption.

The first question.
Is there anything wrong with using GDP and all the components (C, G, I, X, M), which of course will give R2 = 1, if I'm only interested in IRF.

Second question.
Is there a way to analyse multiple simultaneous shocks e.g. shock in exports and consumption at the same time?

  • $\begingroup$ Regarding the second question, I think you can. Do you see any particular obstacles to that? $\endgroup$ – Richard Hardy Mar 23 '20 at 14:12
  • $\begingroup$ I don't see any obstacle per se but maybe I wasn't clear enough. The reason I ask is that in most statistical software you can analyse IRF but (I think) only impulse in one variable at a time e.g. reponse of GDP to C or response of GDP to X. I would like to see the response in GDP to C and X. Could I add those two reponses? $\endgroup$ – Viðar Ingason Mar 23 '20 at 14:30
  • 1
    $\begingroup$ I guess you would need to program that manually. But that should not be too difficult to do (though what is easy for one person may be difficult for another). $\endgroup$ – Richard Hardy Mar 23 '20 at 15:30

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.