Hi, I did as you said Artur and it works perfectly. Grazie per l'aiuto. Guido ----- Messaggio originale ----- Da: artur <ab.news(a)laposte.net&gt; A: Gretl list <gretl-users(a)lists.wfu.edu&gt; Inviato: Domenica 2 dicembre 2007, 15:26:42 Oggetto: Re: [Gretl-users] Forecasting guido mastrolilli a écrit : It's ok Guido & wellcome to Gretlmania! All you have to do is to include the observations for which you need the forecast in the data file before estimating the OLS. For example, if your regression use data is spanning from T1 to T2 and the "out of sample" forecast will be done for data from (T2+1) to T3, the data file you load in Gretl must include the whole data, from T1 through T3. Just before you proceed with the OLS regression switch on the subsample T1 to T2 what leaves room for out of sample forecasting. Buona fortuna, e facci sapere Artur _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users ___________________________________ L'email della prossima generazione? Puoi averla con la nuova Yahoo! Mail: http://it.docs.yahoo.com/nowyoucan.html