Dear May,
Thanks a lor for your reply. I understand the point for the case of
continuous variables. But does this still apply for dummy variables?
In my situation I have the marginal effect for dummy variables being
greater than 1.
Thanks again,
Radu
On 10/6/05, May Boggess <[email protected]> wrote:
> On Thursday afternoon Radu wrote:
>
> > I'm interested in the marginal effects so I'm using
> > the -dprobit- command. And I'm puzzled because the coefficients on
> > some of my explanatory dummy variables are greater than 1. I thought
> > that the coefficients represent the change in probability associated
> > with moving from dummy = 0 to dummy = 1, so they couldn't possibly be
> > greater than 1.
>
> A marginal effect for a continuous variable is a slope and so can
> be greater than 1. Here is an FAQ on the topic that explains
> the issue in detail:
>
> http://www.stata.com/support/faqs/stat/mfx_size.html
>
> -- May
>
> [email protected]
> *
> * 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/
>
*
* 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/
