I am unable to replicate some of Scott Merryman's estimates using his
code. I get that MEM is equal to AME, which does not make sense. I am
using Stata/SE 9.2 on Windows XP, SP2. Perhaps I am missing something
obvious?
Here's my output:
. use "http://www.stata-press.com/data/r9/nhanes2f.dta",clear
. qui svy: probit heartatk height weight female black age highbp
. mfx
Marginal effects after svy:probit
y = Pr(heartatk) (predict)
= .01018915
------------------------------------------------------------------------------
variable | dy/dx Std. Err. z P>|z| [ 95% C.I. ] X
---------+--------------------------------------------------------------------
height | -.0002103 .00018 -1.18 0.237 -.000559 .000139 168.465
weight | .0001798 .00008 2.21 0.027 .000021 .000339 71.9113
female*| -.0127414 .00357 -3.57 0.000 -.019744 -.005739 .520313
black*| -.0006893 .00221 -0.31 0.756 -.005028 .00365 .095637
age | .0011518 .00008 13.93 0.000 .00099 .001314 42.2387
highbp*| -.0033818 .00176 -1.93 0.054 -.006824 .00006 .105917
------------------------------------------------------------------------------
(*) dy/dx is for discrete change of dummy variable from 0 to 1
. margeff, at(mean) dummies(female \ black \ highbp)
Marginal effects at means on Prob(heartatk==1) after probit
------------------------------------------------------------------------------
heartatk | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
height | -.0006499 .0005384 -1.21 0.227 -.0017052 .0004053
weight | .0005557 .0002374 2.34 0.019 .0000904 .0010209
female | -.0380105 .0091002 -4.18 0.000 -.0558465 -.0201744
black | -.0021512 .0069897 -0.31 0.758 -.0158508 .0115484
age | .003559 .0003089 11.52 0.000 .0029536 .0041644
highbp | -.0109917 .0060366 -1.82 0.069 -.0228231 .0008398
------------------------------------------------------------------------------
. margeff
Average marginal effects on Prob(heartatk==1) after probit
------------------------------------------------------------------------------
heartatk | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
height | -.0006499 .0005384 -1.21 0.227 -.0017052 .0004053
weight | .0005557 .0002374 2.34 0.019 .0000904 .0010209
female | -.0373506 .0092254 -4.05 0.000 -.0554321 -.0192692
black | -.0021513 .0069901 -0.31 0.758 -.0158517 .0115491
age | .003559 .0003089 11.52 0.000 .0029536 .0041644
highbp | -.0111666 .0061382 -1.82 0.069 -.0231972 .000864
------------------------------------------------------------------------------
*
* 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/
