I agree. Fitted values is the most appropriate term for within sample and perhaps predicted where the input values in the model differ from the observed ones. Forecast should be retained for out of sample period. But then I didn't write the excellent Gretl package and GUI. But the forecast option does provide CIs, whereas the view of fitted values and errors does not. So I can live with the Forecast misnomer covering both in and out of sample .

If only it didn't join different Panel Unit observations together in the graph option which is where I originally came in !

Brian

On Thu, 29 Feb 2024, 10:37 Sven Schreiber, <sven.schreiber@fu-berlin.de> wrote:

Am 28.02.2024 um 22:15 schrieb Riccardo (Jack) Lucchetti:
> On 28/02/2024 21:59, Brian Revell wrote:
>> Jack, just to clarify, my query was about using the forecast option
>> with a panel dataset, for forecasting within the sample time period
>> range in each panel unit, in which the forecast is the fitted
>> value. The graphics function treats contiguous observations in
>> adjacent panel units and plots them as such when calling up Analysis;
>> Forecast from a pooled OLS model with DVs, thereby making the graph
>> line for the dependent variable continuous. The forecast function is
>> not available when specifying FE or RE Panel models.
>
> I'm aware of that, Brian, thanks.

To be honest, I had forgotten that in-sample predictions are also
subsumed under the term "forecast" in gretl, and I find it a bit
unfortunate. But I just checked the introduction to the Forecasting
chapter in the guide, and it is quite clear about it, so I guess I just
have to get used to it again.

thanks

sven

_______________________________________________
Gretl-users mailing list -- gretl-users@gretlml.univpm.it
To unsubscribe send an email to gretl-users-leave@gretlml.univpm.it
Website: https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/