Panel VAR

[samijo]

Hi,

Is it possible to estimate a Panel VAR in EViews? If for example I wanted to use the grunfeld standard data and run a VAR of the the three variables (I, K & M) allowing for fixed effects; is that possible?

Thanks,

[EViews Gareth]

EViews doesn't have any panel VAR procedures built in.

[startz]

EViews doesn't have any panel VAR procedures built in.

But since a VAR is just OLS, you may be able to get what you need by running panel LS and explicitly writing out the lags on the right hand side.

[Gene]

EViews doesn't have any panel VAR procedures built in.

But since a VAR is just OLS, you may be able to get what you need by running panel LS and explicitly writing out the lags on the right hand side.

Using OLS to estimate panel-VAR is one option. However, the literature suggests alternative ways for panel-VAR estimations. These alternatives are not, yet, available in EViews. It's not clear how much the OLS method deviate from the alternative methods.

[devisetyo]

EViews doesn't have any panel VAR procedures built in.

But since a VAR is just OLS, you may be able to get what you need by running panel LS and explicitly writing out the lags on the right hand side.

i don't know how to write "the lags on the right hand side". could you please tell me how is my equation must be if I have variables U and E, and I want to try lag 1, 2, and 3.
E is Entrepreneurial activity, and U is unemployment. My paper is want to know the dynamic relationship between those two variables.

your answer will very helpful. If my question is not really clear, I'm ready to be asked :

Thanks :)

[startz]

The lag of U is written U(-1), etc.