On 7/26/06, Richard Williams <Richard.A.Williams.5@nd.edu> wrote:
I just tried the following - given that margeff and mfx produce the
same results, I am guessing you are ok (but I can't guarantee it).
use "http://www.stata-press.com/data/r9/nhanes2f.dta";
svy: probit heartatk height weight female black
margeff, at(mean)
mfx
margeff, at(mean) model(probit)
I agree about the simple probit option. However,
margeff, at(mean) model(svy: probit)
yields a fairly different result.
Look at how the output is labeled though. It should probably return
an error when you do that, but instead it apparently thinks you
specified the -eform- option.
