0
$\begingroup$

I am estimating a VAR-model with three (anual) variables: GDP, Unemployment rate and Inflation. (57 observations)

Inflation and GDP are I(1), so I want to use them as growth rates in the model. But the growth rate of the unemployment is not stationary. I tried the first difference of the growth rate, which is indeed stationary.

Now my question:

Can I estimate a Var-model that includes two growth rates and one first difference of a growth rate?

PS: I tried to calculate the growth rate of the growth rate of unemployment, but as the growth rate of unemployment has values that are 0, the growth rate of the growth rate can not be calculated.

My goal is to interpret the model with IRF and granger-causality.

Thanks in advance.

Improve this question
$\endgroup$
3
  • $\begingroup$ I think you could benefit from some macroeconomic theory in addition to purely statistical considerations. $\endgroup$
    – Richard Hardy
    Commented Jun 18, 2021 at 18:57
  • $\begingroup$ Could you please specify? $\endgroup$
    – Joe94
    Commented Jun 18, 2021 at 19:51
  • $\begingroup$ I am not a macroeconomist, but if I needed to investigate the question, I would look at the theory regarding what orders of integration the different variables may have and whether they should be cointegrated in any form. $\endgroup$
    – Richard Hardy
    Commented Jun 19, 2021 at 6:24

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.