reinclude add-factors

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

Moderators: EViews Gareth, EViews Moderator

xprimexinverse
Posts: 39
Joined: Fri Sep 18, 2015 11:41 am
Location: Dublin, Ireland
Contact:

reinclude add-factors

Postby xprimexinverse » Fri Dec 11, 2020 4:07 pm

Hi,

The EViews help doc says that when we use the reinclude model proc it generates an add factor for each variable. However, the add factors that are created using reinclude do not appear in the workfile. My hunch tells me that the add-factors do exist, but that they're hidden from view.

Q: Can someone confirm that this is what happens when using reinclude?

Q: Would it be best if the _a variables were automatically generated in the workfile when using reinclude?

The specified variables are removed from the current active scenario’s exclude list, and generates an add factor for each variable so that the solution for the current scenario remains unchanged.


On my hunch... I think I can detect that the add factors do exist somewhere, but it's difficult for me to explain how. Let me try. It basically boils down to me running identical simulations (with and without reinclude) and then comparing the deviations from makegraph(c,d) to my hand-made deviations (from e.g. series deviation=scenario1-scenario2). My hand-made deviations are 0, but they're positive in the makegraph. This detail is not important to my question, just some background information I've gathered to try to understand the inner workings of the proc.

Cheers,
Graeme

(Note: EViews 10+ Feb 5 2020 Build)

EViews Matt
EViews Developer
Posts: 560
Joined: Thu Apr 25, 2013 7:48 pm

Re: reinclude add-factors

Postby EViews Matt » Mon Dec 14, 2020 11:44 am

Hello,

The created add factors should be present and visible in the workfile as usual, assuming there aren't any view filters active. As a simple test, if you run the program below, do you see the add factor series y_a in he created workfile?

Code: Select all

create u 10
series x = rnd
series y = @trend + 2
series z = 1
model m
m.append x = y
m.append y = z
m.exclude y
m.solve
m.reinclude y
m.solve


Return to “Models”

Who is online

Users browsing this forum: No registered users and 9 guests