Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: perform tests with marginal effects

 From Andreas Fagereng To statalist@hsphsun2.harvard.edu Subject st: perform tests with marginal effects Date Mon, 21 Jan 2013 13:03:02 -0500

```Dear Listers,

I run a probit of a dummy variable on the LHS and two dummy variables
on the RHS.
To get the marginal effects I use the mfx.
I then want to test wether the marginal effect of landsdel1 = e.g. 0.4
When I do this however stata seems to take the coefficients form the
probit regression and not the marginal effects.

How do I make the marginal effects testable?

Best,
Andreas

------------

probit landsdel5y1 landsdel1 landsdel2

Iteration 0:   log likelihood = -661.07518
Iteration 1:   log likelihood = -587.22295
Iteration 2:   log likelihood = -586.96204
Iteration 3:   log likelihood = -586.96204

Probit regression                                 Number of obs   =        955
LR chi2(2)      =     148.23
Prob > chi2     =     0.0000
Log likelihood = -586.96204                       Pseudo R2       =     0.1121

------------------------------------------------------------------------------
landsdel5y1 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
landsdel1 |    1.13818   .0979193    11.62   0.000     .9462612    1.330098
landsdel2 |   .0667942   .1894963     0.35   0.724    -.3046118    .4382002
_cons |  -.2772226   .0515024    -5.38   0.000    -.3781655   -.1762797
------------------------------------------------------------------------------

. mfx, at(landsdel1 =0 , landsdel2=0)

Marginal effects after probit
y  = Pr(landsdel5y1) (predict)
=   .3908046
------------------------------------------------------------------------------
variable |      dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
---------+--------------------------------------------------------------------
lands~l1*|   .4145645      .03028   13.69   0.000   .355215  .473914         0
lands~l2*|   .0258621      .07386    0.35   0.726  -.118891  .170616         0
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1

.
*
*   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/
```