Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Comparing interaction terms in simultaneous equations


From   Steve Binggeli <binggeli.steve@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Comparing interaction terms in simultaneous equations
Date   Sat, 27 Oct 2012 10:55:55 -0500

Hello,

I am using cmp to perform simultaneous regressions for panel data.
I have two equations, same model, but different outcomes:
The two dependent variables are "job desirability" for boys and girls.
Hence, it is the same measure but repeated for two different targets.
The predictors are gender, ethnicity (dummy, 0-1) and their interaction term.

In this particular case, I was wondering whether comparing the two
interaction terms across equations (with a Wald test) could be
considered as a test of a three-way interaction (gender x ethnicity x
gender of children (boys vs girls)? If not, what would the result of
this test mean?

Ideally, I would also like to get the predicted margins from the two
interactions terms (across equations) to compare them.
Do you know how I could obtain those margins?

Here is the cmp command that I am using:

 xi: cmp (girljob = gender##ethnicity  ///
  i.parents  ses || participantid:) ///
 (boyjob =  gender##ethnicity  ///
  i.parents  ses || participantid:), cluster(participantid) ind( 1 1)
redraws(101) difficult

Thanks for your help!

Best.

Steve
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index