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

# RE: st: RE: RE: what does it mean the default for -mfx- is discrete when it is evaluated at the mean?

 From "Martin Weiss" To Subject RE: st: RE: RE: what does it mean the default for -mfx- is discrete when it is evaluated at the mean? Date Thu, 3 Jun 2010 11:16:04 +0200

<>

You can try all of this yourself. As long as you use default settings (first
loop below), Stata treats the dummy covariate as a discrete variable, in
accordance with intuition, and the point at which you evaluate it does not
change the marginal effects. It is only when you tell it to treat the dummy
as a continuous variable (second loop) that the marginal effect changes. The
other covariates are held at their mean throughout, btw.

***********
sysuse auto, clear
gen byte dummy=rep78<4
logit  foreign gear_ratio turn length dummy
forv i=.1(.4).9{
mfx compute, varlist(dummy gear_ratio) dydx at(mean dummy=`i')
}
forv i=.1(.4).9{
mfx compute, varlist(dummy gear_ratio) dydx at(mean dummy=`i') nodisc
}
***********

HTH
Martin

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nirina F
Sent: Donnerstag, 3. Juni 2010 04:18
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: RE: RE: what does it mean the default for -mfx- is discrete
when it is evaluated at the mean?

Thank you very much for the responses.
May be : What you mean is that the rest of the variables are evaluated
at the mean but the independent dummy variable is evaluated at the
value 1 (instead of for example the mean at .37)
Nirina

On Tue, Jun 1, 2010 at 5:39 PM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>
> <>
>
> This was a knee-jerk reaction of course, since in a linear regression the
> marginal effects would always be constant.
>
> Still, what is the problem with saying that the marginal effect displayed
is
> for going from zero to one, i.e. the only move that a dummy can possibly
> make? The fact that the share of ones in your sample is 37% does not
change
> this fact, does it?
>
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Martin Weiss
> Sent: Dienstag, 1. Juni 2010 23:27
> To: statalist@hsphsun2.harvard.edu
> Subject: st: RE: what does it mean the default for -mfx- is discrete when
it
> is evaluated at the mean?
>
>
> <>
>
> The point of evaluation does not seem to matter at all in this code,
though:
>
>
> ***********
> sysuse auto, clear
> reg price weight length foreign
> mfx compute, dydx at(mean foreign=0)
> mfx compute, dydx at(mean foreign=1)
> mfx compute, dydx at(mean foreign=.2)
> mfx compute, dydx at(mean foreign=.5)
> mfx compute, dydx at(mean foreign=.8)
> ***********
>
>
> HTH
> Martin
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Nirina F
> Sent: Dienstag, 1. Juni 2010 16:09
> To: statalist@hsphsun2.harvard.edu
> Subject: st: what does it mean the default for -mfx- is discrete when it
is
> evaluated at the mean?
>
> Hello,
> I have a dummy for dependent and main independent variable. I am using
> ivprobit.
> When I try to get the marginal effects, mfx  shows you that the
> default setting is 'discrete', that is, evaluate the marginal effect
> of a dummy going from 0->1. But it is evaluated at the mean (and under
> x it shows 0.37 for the independent variable) so I am a little bit
> confused about the interpretation of the marginal effect.
> Thank you for your help.
> Nirina
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/