[Gretl-users] Re: multi-step-ahead-forecasting
Yes, now it is ok, thank you very much
it gives error without appending --dynamic option
Cheers,
On Thursday, January 7, 2021, 10:19:48 PM GMT, Sven Schreiber <svetosch(a)gmx.net>
wrote:
Am 07.01.2021 um 23:07 schrieb Burak Korkusuz:
Thanks for your reply,
maybe, that one is clearer. This is for one-step ahead forecasting. I am
trying to generate 5 (and 22) -step-ahead out-of-sample forecasts using
the equation (ols djclose const djclose(-1)) starting of the sample
period 1+i – 100+i. However, the equation is estimated recursively
every time and forecasted every time using the new equation fitted values.
open djclose.gdt
set verbose off
series frcst = NA
loop i=1..200 -q #out-of-sample (10 observations)#
smpl 1+i 100+i #initial sample (5 observations)#
ols djclose const djclose(-1)
fcast 101+i 101+i #one-step-ahead-rolling-windows-forecasting#
frcst[101 + i] = $fcast
endloop
Well inserting the FC horizon specification should still work here.
Since you're forecasting out-of-sample in every step the forecast should
be dynamic by default (check the fcast reference), but if you want to be
absolutely sure you may append the --dynamic option to the fcast line.
cheers
sven
_______________________________________________
Gretl-users mailing list -- gretl-users(a)gretlml.univpm.it
To unsubscribe send an email to gretl-users-leave(a)gretlml.univpm.it
Website: https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/
Attachments:
- attachment.htm (text/html — 3.4 KB)