EViews.com

Dear Eviews-community,

The EViews 10 Feature list says that “Estimation with AR errors using nonlinear least squares on a transformed specification” should be possible. Unfortunately I wasn’t able to figure out how to program/estimate it. Could someone help me with this, especially in a panel context?

If I just type “eq_name.ls(cov=cxsur) y c x ar(1)” I get an AR-term in the cross-section but not AR-errors.

Thank you very much for your help

I get an AR-term in the cross-section but not AR-errors.

What does that mean?

Dear Gareth,
thanks for your quick reply.
With "AR-term in the cross-section" I wanted to say that a lagged dependent variable is included in the regression. Actually, I'd like to estimate pooled GLS with a panel-wide AR(1) correction where the coefficient of the AR(1) process is common to all the panels.

The corresponding STATA-code would look like this: xtpcse y x , correlation(ar1) hetonly

Would be very nice, if you could help me with that.