Re: st: Partial F Test on a Multiple Imputed Set

[Richard Williams <richardwilliams.ndu@gmail.com>]

You are testing the hypothesis that the coefficients for the dummies 
are equal in magnitude but opposite in sign. Are you sure that is 
what you want, as opposed to just testing that both coefficients equal 0, e.g.

test D1 D2

This will work with mi test.

But, if you really want to test your hypothesis as stated, see

help mi test

It includes examples for more complicated forms of hypotheses.

At 10:09 AM 5/27/2013, Eduardo Moreira wrote:
> Dear all,
>
> I am sorry if this has been asked before, but I searched and 
> couldn't find the answer to this question.
>
> Say you want to compare 2 regression models on the same dependent 
> variable Y. The second model contains 2 dummy variables, D1 and D2.
>
> - Model A: Y = b0 + bX
> - Model B: Y = b0 + bX + bD1 + bD2
>
> usually what I would do is:
>
> regress Y X D1 D2
> test D1 + D2 == 0
>
> So far, so good, right?
>
> The problem I run into is when I try to do the same with a mulitple 
> imputed set. I run the following commands:
>
> mi estimate: regress Y X D1 D2
> mi test D1 + D2 == 0
>
> The regression works fine, but test part doesn't. It gives the 
> following error message: "=exp not allowed".
>
> Is there some way to perform this Partial F-test on a MI set?
>
> Thanks,
> Eduardo Moreira
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/