Problem with dummy variable in PMG model

[Osama Elsalih]

Hello,

I'm running PMG model with one dummy veritable of (0 or 1) (xtpmg d.y d.x1 d.x2 d.x3 d.x4 DUM , lr(l.y l.x1 l.x2 l.x3 l.x4 DUM ) ec(ECT) replace pmg),

Unfortun stata show this message (numerical derivatives are approximate flat or discontinuous region encountered)

So, please I do need your help to correct this problem.

Thanks in advance

[Carlo Lazzaro]

Osama:
in all likelihood you should revise your model specification.

[Osama Elsalih]

Thanks Carlo,

Actually I couldn't understand what do you mean by revise my model. What exactly should I do, I'm not really professional!

Thanks
Osama

[Carlo Lazzaro]

Osama:
you should add one predictor at time and see when Stata starts giving the error message.
As an aside, I cannot believe that you are not more than skilled in econometrics if you embark yourself in such a complex and methodologically demanding regression model.

[Osama Elsalih]

Carlo

I got what you mean now !!! Yes I did that, I dropped the variables, variable by variable. Unfortunately the error occur when I add the dummy variable which I do need it to in the model.

Also,I tried to run the model by OLS, FE and RF as a test, the model work well with dummy !!!!

Osama