Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Marginal Effects


From   MERSKY@Waisman.Wisc.Edu
To   statalist@hsphsun2.harvard.edu
Subject   st: Marginal Effects
Date   Thu, 27 Apr 2006 13:08:28 -0500 (CDT)

Statalist members,

I am hoping to obtain syntax that will allow me to obtain the proper
adjusted means for models in which I ask for marginal effects in probit
regression.  STATA often suppresses the true means, although it does
preserve the true proportional mean differences.  I believe that Cohen
et al. (2002) recommend using the unadjusted mean for the reference
group and adding the adjusted marginal effect for the independent
variable of interest. The problem with this approach, of course, is that
it results in smaller proportional differences between groups. I would
like to avoid this approach. Is there another way that I can get the
"true" adjusted group means (marginal effects) while maintaining the
appropriate percentage-point differences?

I have pasted an abridged version of my syntax below for reference.
Many thanks in advance,

Joshua Mersky


probit violdelr anyprek anyfol can011b etc....
mfx compute,dydx nose at (mean can011b = 0)
*
*   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