## RE: Authomatic ARIMA Forecasting

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

eviewsuser10
Posts: 14
Joined: Thu Oct 15, 2020 3:27 am

### RE: Authomatic ARIMA Forecasting

I conducted automatic ARIMA Forecasting. Eviews provides a comparison graph between observed and forecasted values. I just wonder whether Eviews provides confidence intervals for forecasted values. I would appreciate any help.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12908
Joined: Tue Sep 16, 2008 5:38 pm

### Re: RE: Authomatic ARIMA Forecasting

Not from the automatic forecasting, but you can create an equation object from the forecast (one of the options on the dialog), and then forecast from that equation, which will give you the full set of forecasting options, including standard errors.

eviewsuser10
Posts: 14
Joined: Thu Oct 15, 2020 3:27 am

### Re: RE: Authomatic ARIMA Forecasting

EViews Gareth wrote:create an equation object from the forecast (one of the options on the dialog), and then forecast from that equation.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12908
Joined: Tue Sep 16, 2008 5:38 pm

### Re: RE: Authomatic ARIMA Forecasting

eviewsuser10
Posts: 14
Joined: Thu Oct 15, 2020 3:27 am

### Re: RE: Authomatic ARIMA Forecasting

Dear Gareth,

Following your suggestions, I conducted automatic ARIMA forecasting with a seasonal dummy variable and a lagged dependent variable, which gave me ARIMA (0, 1) as the best fitted model. Thus, I used ARIMA (0, 1) with the same model specification with the previous one to forecast with confidence intervals (using static forecasting). Please see the attached. I compared the graphs created by the automatic ARIMA forecasting and by the equation of ARIMA (0,1). I just wonder why two graphs are so different. The first one captured well the volatility of the series. The second one did not. Should both graphs be the same? Any help would be appreciated.
Attachments
Doc1.docx

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12908
Joined: Tue Sep 16, 2008 5:38 pm

Its hard to say.