You might find Jonah B. Gelbach's margfx program helpful - it will calculate
the mean marginal effects for probit and logit models.
You can find a link to his homepage from the Stata homepage at
http://www.stata.com/links/resources2.html
Scott Merryman
----- Original Message -----
From: "Elisabeth Simantov" <[email protected]>
To: <[email protected]>
Sent: Friday, July 19, 2002 10:58 AM
Subject: st: FW: marginal effects after probit for each observation in the
data set
>
> > -----Original Message-----
> > From: Elisabeth Simantov
> > Sent: Friday, July 19, 2002 9:55 AM
> > To: '[email protected].'
> > Subject: marginal effects after probit for each observation in the
> > data set
> >
> > I am running a probit regression with 19 independent variables, 510
> > observations. The canned version of marginal effects calculates around
the
> > mean. I would like to obtain a matrix that contains marginal effects for
> > each variable for each observation and then take a mean over all
> > observations for each marginal effect.
> > The formula for the derivative looks somewhat like this:
> >
> > <<...OLE_Obj...>>
> >
> > where f is the standard normal density function. This is calculated for
> > each observation using its value for the vector of independent
variables,
> > X, together with the estimated probit coefficients, and is then averaged
> > across the full sample.
> >
> > I am having a problem translating this to STATA language. I hope that
> > someone can help me. I am not a proficient programmer, so please be very
> > explicit.
> >
> > Thank you in advance.
> > Elisabeth Simantov
> >
> > Elisabeth Simantov, PhD
> > Research Associate
> > Center for Home Care Policy and Research
> > Visiting Nurse Service of NY
> > 107 East 70th Street
> > New York, NY 10021
> >
> > email: [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/
