Replace 'actuals' with 'baseline'

For technical support, tips and tricks, suggestions, or any other information regarding the EViews model object.

Moderators: EViews Gareth, EViews Moderator

HollEviews
Posts: 15
Joined: Wed Mar 12, 2014 6:52 am

Replace 'actuals' with 'baseline'

Postby HollEviews » Wed May 21, 2014 3:39 am

Please can you tell me how to replace the Actuals with the Baseline data? I have a ragged edge dataset, and want to perform a Fit solution up to the latest point in the dataset, so I need to fill in the actuals series with forecast values for the missing observations.
Thank you!

EViews Glenn
EViews Developer
Posts: 2671
Joined: Wed Oct 15, 2008 9:17 am

Re: Replace 'actuals' with 'baseline'

Postby EViews Glenn » Wed May 21, 2014 11:51 am

Are the missings in the series only for periods beyond the latest point? Are you sure you want to modify the actuals, or do you want to splice the actuals and the forecasts into a new series?

HollEviews
Posts: 15
Joined: Wed Mar 12, 2014 6:52 am

Re: Replace 'actuals' with 'baseline'

Postby HollEviews » Wed May 21, 2014 12:46 pm

In my 'actuals' I have several data series, with end points between 2010 and 2013. I want to be able to run a "Fit" solution to 2013, so I need to fill in the missing values between 2010 and 2013 somehow. I don't particularly want to overwrite my actuals data, but is there a way to use the baseline output (which fills the 2010-13 values with equation projections) to run the Fit solution? As a temporary solution I've saved to a separate workfile, deleted all of my actuals, and renamed the _0 series as the actual names. That works okay, but I was hoping there was a slightly neater method. Many thanks for any suggestions!

EViews Glenn
EViews Developer
Posts: 2671
Joined: Wed Oct 15, 2008 9:17 am

Re: Replace 'actuals' with 'baseline'

Postby EViews Glenn » Wed May 21, 2014 2:09 pm

Are the series that you want to fill the exogenous variables? And I'm still need to know whether the only missings are after the endpoints. And is there a reason you want to use the fit rather than another scenario?

HollEviews
Posts: 15
Joined: Wed Mar 12, 2014 6:52 am

Re: Replace 'actuals' with 'baseline'

Postby HollEviews » Thu May 22, 2014 2:07 am

All the variables in my model are endogenous. And yes, the missing values are only after the endpoint of each series. I'm using the Fit solution, because I want to assess the historical errors on each equation, so that I can set the add factors for the forecast after 2013. (which I'll run as a dynamic solution).

EViews Glenn
EViews Developer
Posts: 2671
Joined: Wed Oct 15, 2008 9:17 am

Re: Replace 'actuals' with 'baseline'

Postby EViews Glenn » Fri May 23, 2014 6:56 am

Code: Select all

y = @recode(y=na, y_0, y)

sets y to y_0 if it's missing and leaves the value in place otherwise. If you have a lot of endogenous series, I'd put the names in a loop or put all of the series in a group and iterate over the series in the group.

HollEviews
Posts: 15
Joined: Wed Mar 12, 2014 6:52 am

Re: Replace 'actuals' with 'baseline'

Postby HollEviews » Thu May 29, 2014 6:44 am

Many thanks!!


Return to “Models”

Who is online

Users browsing this forum: No registered users and 13 guests