I just randomly wrote my code as an example for that dataset to be clear. Yes, I want to have a rolling initial sample with a constant. Actually in my real dataset I use the lagged variables of my dependent variable (it is a time series regression). I can consider only the first lag of dependent variable and constant in the equation.

On Thursday, January 7, 2021, 08:30:36 PM GMT, Sven Schreiber <svetosch@gmx.net> wrote:

Am 07.01.2021 um 20:23 schrieb Burak Korkusuz:

> Hi,
> Below is an example for one-step ahead forecasting that I use the first
> 5 obs. for initial sample and then forecast one-step-ahead by rolling ahead.
> My question is that how should I change my code to get 5-step-ahead
> rolling forecast.
> Thanks,
>
> open denmark.gdt
> set verbose off
> series frcst = NA
> loop i=1..20 -q #out-of-sample (20 observations)#
> smpl 1+i 5+i #initial sample (5 observations)#
> ols LRM const LRY
> fcast 6+i 6+i #one-step-ahead-rolling-windows-forecasting#
> frcst[6 + i] = $fcast
> endloop

>

Hi, two questions first: Do you really want to have a rolling sample
with a constant sample length of 5 obs? Or rather move only the end obs?
Secondly, do you really want to have no lagged variables in your
equation? While technically feasible, a "forecast" from an equation
where all regressors except the constant are unknown for the next period
doesn't give anything really meaningful.

cheers
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/