Hello,
I am wondering if its possible to estimate seemingly unrelated regression (SUR) system with error-correction process. It would kind of be like a VECM but with asymmetric lags in the individual equations. I have tried introducing an ECT into my SUR, but I am not entirely sure how to go about estimating this in RATS. I believe that if you introduce an ECT into a SYSTEM, you have to use a symmetric MODEL with the same number of lags.
Thanks!