st: xtpmg - residual calculation

[andreas.pondorfer@ifw-kiel.de]

Dear Statalist,

I am using the stata command xtpmg written by Blackburne and Frank to
estimate nonstationary heterogeneous panels. To complete my analysis, I
would like to calculate the residuals and test them for cointegration
(Petroni and Kao cointegration tests in Gauss).Unfortunately, there is no
predict command for residuals after the xtpmg command.
Hence, I tried to compute them by my own, using the example sample set
jasa2 (Pesaran et al. 1997, 1999) to get the idea. However, I am not
exactely sure how to fit equation (8) (Stata Journal 7, 2007, p.202) with
the stata output of xtpmg, pmg.
Here are my questions:

-Does phi*theta_zero_i in the bracket of (8) represent the _cons in the
short run dynamics? Or am I completely wrong?

-How can I compute the residuals in the most simple way?

Thank you for your help.

Yours sincerely,

Andreas Pondorfer







*
*   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/