Allin (and friends),
someone here asked to me whether we may do forecasts "out of the sample" with
gretl, this means calculate fitted values for the dependent variable that
corresponds to new values of the independent variables. I can see that, once
a model has been estimated, in the output window we may select "/Model
data/Forecasts with standard errors" but we may only select here observations
inside the sample.
I think it would be interesting to have a new entry in the /Model data/ menu
that allows calculating some forecasts and confidence intervals of Y for new
values of X. ¿Could you think about it?
--
Ignacio Díaz-Emparanza
Dpto. de Economía Aplicada III (Econometría y Estadística)
UPV-EHU
http://www.bl.ehu.es/~etpdihei/