Re: [Gretl-users] Forecast ARIMA
On Thu, 12 Aug 2010, [koi8-r] Александр Рогальский wrote:
During the simulation the following problem occured.
This problem concerns the construction of confidence intervals.
Having developed the ARIMA model (Time Series->ARIMA, using the
Exact LikelyHood method) I made a forecast of the process in а
five-year term.
The predicted values were calculated correctly, but all
confidence intervals had the same size (but the must expand with
increasing period of pre-emption).
What is the reason of it?
I don't know without seeing your model and your data.
I'm not able to replicate a non-increasing confidence interval
from an ARIMA model in examples I've tried. If your data set is
not too large you could post it here (and also specify precisely
the model you are estimating).
Allin Cottrell