st: standard error of forecast, weighted regression

["Carlo Fezzi" <C.fezzi@uea.ac.uk>]

Dear all,
�
I fitted a linear regression using analytical weights (with cluster for
residuals), and now I am trying to make some forecasts with predict.
�
Unfortunately, I am not able to obtain the standard error of forecast, since
the option stdf is not allowed with weighted data. According to the Stata
manual the reason is that �this is not a statistically well-defined
concept�, but no specific reference is given to explain why this is the
case.
�
I would be really grateful if anybody could suggest a function or an .ado
file I can use to obtain this result, or cast some light on the reason why
this is not a �good thing to do�.

Thanks a lot for your help,

Carlo Fezzi
University of East Anglia



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/