st: residuals after pmg regression

[andreas.pondorfer@ifw-kiel.de]

Dear Statalist,

I am using the xtpmg command to estimate a large N and large T panel-data
model. Now I´d like to compute the residuals of my regression. My approach
would be to predict the fitted values of the long-run relationsip (i.e.
predict yhat, eq(ec)) and use those results to calculate the residuals
(y-yhat). I need the residuals for cointegration tests (Kao, Pedroni).

Has someone experience with the pooled mean group and dynamic fixed effect
estimator and can help me out? Is this the correct way to compute the
residuals for the long-run relationship?

Thanks,
andy

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/