Forecasting from a System

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

27npg
Posts: 29
Joined: Thu Jul 25, 2019 9:31 am

Forecasting from a System

Postby 27npg » Wed Sep 04, 2019 5:32 am

Hi – I have been trying to run forecasts using the output of a system. I am aware that this is not possible through the ‘forecast’ option and instead users are required to Proc->Make Model-> and forecast here through the ‘solution sample’ window. However, whenever I do this I keep getting the following error message.

Scenario: Baseline
Solve begin 13:28:43
2019Q2 NA generated for NFC
Solve terminated - Unable to compute due to missing data in "NFC = @COEF(1) + @COEF(2) * INTEREST_RATE + @COEF(3) * DEPOSITS …


This does not happen when I forecast in this way let’s say using a VAR or VECM. My estimated model is from 1986Q1 TO 2019Q1 and am trying to forecast 2019Q2-2020Q4. Any ideas?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13317
Joined: Tue Sep 16, 2008 5:38 pm

Re: Forecasting from a System

Postby EViews Gareth » Wed Sep 04, 2019 6:19 am

Sounds like you don’t have data for the right hand side variables over the forecast period.
Follow us on Twitter @IHSEViews

27npg
Posts: 29
Joined: Thu Jul 25, 2019 9:31 am

Re: Forecasting from a System

Postby 27npg » Wed Sep 04, 2019 6:45 am

Hi Gareth - I have data for all variables up until 2019q1. This allows me to forecast su9ng VAR/VECM. Is this different when using a system? Do I need future values to forecast my dependent variable?

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Forecasting from a System

Postby startz » Wed Sep 04, 2019 7:03 am

A VAR only uses lags on the right. It uses these to generate the current value, which then becomes the required lag the following period. In you model, you need to provide data for all the exogenous right hand side variables.

27npg
Posts: 29
Joined: Thu Jul 25, 2019 9:31 am

Re: Forecasting from a System

Postby 27npg » Wed Sep 04, 2019 7:25 am

Ah of course - how foolish of me. I hadn't included the lags on the RHS. thanks for pointing me in the right direction.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 38 guests