Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: test equality of coefficients from 2 regressions


From   baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: test equality of coefficients from 2 regressions
Date   Thu, 18 Jul 2002 08:27:16 -0400

--On Thursday, July 18, 2002 2:33 -0400 Delane wrote:

I ran 2 regressions (using regress), using the same variables, only
changing the observations. Now I have to test if the 2 coefficients of
the same variable are equal.

I only know how to do this when the coefficients are in the same
regression, or in multiple equations as in mlogit or sureg.

How to test the 2 coefficients in 2 regressions?
This was the subject of a long discussion on the list earlier this year. You cannot perform formal tests when the coefficients are not being estimated in the same model. sureg is one way around this, which since it makes no assumptions about the coefficients, variances, etc. across the equations is workable--but difficult if you have different N's in the two equations. Any test statistic will have to take into account the est.covariance between the two coefficients, and unless they are in the same model, there is no way to calculate that (and surely not appropriate to assume it away).

Kit Baum
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/




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