1
$\begingroup$

I have five independent variables: oil (stationary at level), f (stationary at level), k (stationary after differencing twice) , pd (stationary after differencing), m (stationary after differencing) and dependent variable rc (stationary at level). How could use or write the model on Eviews? Are my inputs displayed in the graph below correct? And could I use vecm?

enter image description here

$\endgroup$
  • $\begingroup$ You should add more information about what you are trying to achieve with your model. Then it may be possible to give a better answer that the one I gave below. $\endgroup$ – mpiktas Dec 9 '14 at 12:14
0
$\begingroup$

Since you say that you have one dependent variable and five independents, then you need to use simple regression and stay away from VAR and VECM.

$\endgroup$
  • $\begingroup$ Thank you mpiktas ...I didn't understand why I don't have to use var and what details do you need to view please? $\endgroup$ – T.G. Zain Dec 9 '14 at 18:14
  • $\begingroup$ You use VAR when there is more than one endogenous (dependent) variable among the variables you have. VAR allows for mutual dependence, e.g. when $rc$ depends on (lagged) $oil$ and $oil$ depends on (lagged) $rc$. There are as many equations in a VAR as there are dependent variables. If only one variable is endogenous and all the other are exogenous (independent), then your VAR will only have one equation and thus will no longer be a VAR but rather a simple regression. $\endgroup$ – Richard Hardy Dec 10 '14 at 1:50
  • $\begingroup$ yes i have two endogenous rc and ris but i thought i have to test each of them in separate way rc and ris are returns for two mutual funds an i want to test if the exogenous variable will cause the endogenous by using Granger causality $\endgroup$ – T.G. Zain Dec 10 '14 at 23:32

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.