EViews.com

I am using seemingly unrelated regression (SUR) to estimate a system of equations with contemporaneous correlation and cross equation restrictions. Is there any possibility to use Newey-West standard erros (HAC) to correct for serial correlation?

I know that I could use GMM/2SLS with the exogenous variables as instruments to implement HAC standard errors if the equations were independent. But how can I perform in my case with contemporaneous correlation and cross equation restrictions? Is there an Add-In I could use or do you have any other suggestions?

I use version 8.

Thanks

There's nothing built in that will do it, unfortunately.

Well, I was hoping you wouldn't say that. Maybe you can implement it in the next version .

Any other ideas how I could proceed?

Thanks

Dear Eviews expert,
I also have the same problem, implementing HAC in SUR. I am using Eviews 8. Can you suggest an approach to do this? Thank you so much!
Regards
Amber