[Gretl-users] Re: forecast interval shading (Re: Re: Package updates (December 2024))
Hi Allin
the script worked fine. I think he issue was that I was hoping to get the
shaded area to apply to pre-forecast fitted values as well as the post
estimation sample estimates, which was why it was unwilling to offer a
shaded area but would give only error bars. If I shorten my estimation
period to leave to reserve years for the forecast, the the shaded area is
graphed for the post estimation sample predicted values.
Problem solved.
Thanks
On Fri, 9 May 2025 at 13:28, Cottrell, Allin <cottrell(a)wfu.edu> wrote:
On Fri, May 9, 2025 at 5:21 AM Brian Revell
<bjr.newmail(a)gmail.com> wrote:
>
> Re 2025a Gretl version
> Sven
> the 2 attached screen grabs clearly show for both OLS and Arima that the
options for the CI are greyed out with effectively a default to range bars.
The "range bars" only situation is supposed to arise when only a
single forecast is possible. Could you try running the following
script and see what you get in the forecast dialog, please?
nulldata 50
set seed 654321
setobs 1 1975 --time
series u = normal()
series y = 5
y = 0.5*y(-1) + u + 0.2*u(-1)
smpl ; 2012
arma 1 1 ; y -w
Allin Cottrell
_______________________________________________
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 — 2.5 KB)