0
$\begingroup$

What is a Redundant Fixed Effects Test? How do I run it in Stata?

I can't find much about it using the help in Stata or by searching on the Web.

Any help will be appreciated.

Thank you

$\endgroup$
1
  • $\begingroup$ To me, the term Redundant Fixed Effects Test seems to simply imply testing whether or not one should include a particular fixed effects term in his/her model. There are numerous resources on the web (and on this forum) regarding variable selection in mixed effects models, hypothesis testing in LME, etc. See for example glmm.wikidot.com's FAQ. If this is not what you want can you please provide more information/references on what you try to achieve? $\endgroup$
    – usεr11852
    Apr 6, 2015 at 5:28

3 Answers 3

1
$\begingroup$

I also don't have enough reputation to either vote up or comment on @marquisdecarabas post, so I'm writing here:

There are two types of tests that you can run.

1) You can compare the pooled regression vs the fixed effects model, by using a simple F-Test or LR test. The FTest is done by Stata with the areg command (the last line of the table).

2) The other test is as marquis said the Hausman test, which you can run with the hausman command (see the help, you need to previously run an xtreg command)

$\endgroup$
0
$\begingroup$

I do not have enough reputation, so I cannot post this as a comment. A cursory glance at both Google and Google Scholar for "redundant fixed effects test" (use the quotation marks in your search to get hits that contain the exact string) produced search results that you may find usable:

Here's another link http://ebooks.narotama.ac.id/files/Introducting%20Econometric%20Financial/Chapter%2010%20Panel%20data.pdf

The term "redundant fixed effects test" is not used in my field. From what I gathered, this test is associated with a software called Eviews. At any rate, based on the search results I linked to above, it seems the term is used to refer to what we call the Hausman test. You should establish first whether the two tests are the same. If they are, search in Stata or Google for the Hausman test in Stata, and you will have no problems figuring out how to do it then.

$\endgroup$
0
$\begingroup$

The redundant fixed effects test in EViews is not what is commonly called the Hausman test. The Hausman test, at least in my research area, refers to a test between fixed and random effects. Eviews implements fixed effects with dummy variables and the redundant fixed effects test in Eviews is simply an F-test if all dummies are equal to zero.

$\endgroup$
1

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.