EViews.com

Bear with me this might be a little long winded as I am completely new to this software.

I am trying to use SUR to estimate this 2 system of equations

sgx = c(1) + c(2)*lse + c(3)*nyse @ ukunemp usunemp
gdppc = c(4) + c(5)*nyse + c(6)*lse @ ukunemp usunemp

and using SUR I got this table

Coefficient Std. Error t-Statistic Prob.

C(1) 568.2343 368.4900 1.542062 0.1329
C(2) 0.369901 0.397356 0.930905 0.3589
C(3) 0.091571 0.119137 0.768619 0.4478
C(4) 13930.48 4276.892 3.257150 0.0027
C(5) 3.170561 1.382767 2.292910 0.0286
C(6) -3.392305 4.611926 -0.735551 0.4674

The first question is this, how can I be sure that eviews has used both ukunemp and usunemp as the IV for the 2SLS estimation before carrying on with the SUR estimation?

Secondly, shouldnt the SUR equation combine the coefficients to form a large matrix so I am able to do GLS estimation? How do I do that in eviews?

Lastly, can I estimate this in SUR:
sgx = c(1) + c(2)*lse @ ukunemploy
sgx = c(3) + c(4)*nyse @ usunemploy

Will there be any specification errors or violation of the SUR model if both the dependent variables are the same? The reason I ask is because I only want ukunemploy to be the IV for the first model and usunemploy to be the IV for the 2nd model, instead of lumping both IVs together in one equation like in the first model.

Thanks for answering :)

1. You seem to be doing 3SLS, not SUR.
2. You can trust EViews to use the instruments you specify. It wouldn't fib to you.