Dear all,
I realized that WLS estimation results in gretl are different from those in Statain the case when observations with weight=0 do exist.
Now, in gretl, the mechanics of the algorithm drops these observationsfrom the estimation process. And indeed, both Stata and gretl report the samenumber of observations (ie. observations actually involved in computing) with gretlgiving addtional info on dropped observations.
Consequently, gretl doesn't provide postestimation values on these droppedobservations (uhat, yhat)...but Stata does!!! And I struggle to understand the logicbehind Stata's proceeding and how it affects the estimates...
I don't know what Stata does there
(although I suspect it is documented), but in principle it seems
easy enough to compute a fitted value yhat even for any obs
(y-X-constellation) which wasn't part of the estimation sample.
I mean that's what we do for prediction, no? Does a (static)
"fcast" work there in gretl?
thanks
sven