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 <cottr...@wfu.edu> wrote: > On Fri, May 9, 2025 at 5:21 AM Brian Revell <bjr.newm...@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@gretlml.univpm.it > To unsubscribe send an email to gretl-users-le...@gretlml.univpm.it > Website: > https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/ >
_______________________________________________ Gretl-users mailing list -- gretl-users@gretlml.univpm.it To unsubscribe send an email to gretl-users-le...@gretlml.univpm.it Website: https://gretlml.univpm.it/postorius/lists/gretl-users.gretlml.univpm.it/
