Redundant variables test and restricted sample

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

Posts: 16
Joined: Sat May 31, 2014 9:09 am

Redundant variables test and restricted sample

Postby chungrak9 » Wed Aug 13, 2014 10:49 am

Hello. I use EViews 8. I have 40 countries for 56 years in the data. Of course, there are many missing values. I did pool estimation and tried to do the redundant variables test, but only saw the following error message:

"Test specifications have different samples due to NAs in original or test series. Reestimate original equation with restricted sample and rerun test."

Would you please let me know how to restrict the sample? I'd like to do a redundant variables test on the country-specific effects. Please refer to the attached files.

"Balance Sample" leaves no observation, thus it's not an option. The same error appears when I checked the "No d.f. correction" box. I know the command "sample" and "smpl", but I don't know what to sample.

Thank you very much.
redundant_var_test_error.jpg (60.31 KiB) Viewed 1892 times
redundant_var_test_estimation.jpg (163.2 KiB) Viewed 1892 times

EViews Glenn
EViews Developer
Posts: 2602
Joined: Wed Oct 15, 2008 9:17 am

Re: Redundant variables test and restricted sample

Postby EViews Glenn » Thu Apr 02, 2015 10:02 am

Unless there is something strange about the data in your workfile, create a group containing

Code: Select all


Call it something like G1. Restimate the equation with a modfied sample that excludes the observations that your unrestricted model would otherwise automatically exclude

Code: Select all

1 56 if @rnas(G1)=0

What's going on is that your test is excluding the lagged value variable so the restricted model using the specified sample has too many observations.

Posts: 3
Joined: Mon Oct 03, 2016 6:58 am

Re: Redundant variables test and restricted sample

Postby kimmy.huong » Mon Apr 16, 2018 7:19 am


I am trying to run redundant variables test for panel data but got the same issue as yours,
Total company: 245, Year: 2007-2015
Dep. variable = FOR
Redundant variable LISTING PERIOD
I created G1= FOR(-1)
at sample i type: 2007 2015 if @rnas(G1)=0

Eviews showed error: "Error in the sample: @RNAS(G1) required a group argument"
Could you please guide me where did I do something wrong?

Many thanks in advance

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 12199
Joined: Tue Sep 16, 2008 5:38 pm

Re: Redundant variables test and restricted sample

Postby EViews Gareth » Mon Apr 16, 2018 7:36 am

Did you make a group called G1?
Follow us on Twitter @IHSEViews

Return to “Programming”

Who is online

Users browsing this forum: No registered users and 4 guests