Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: PARTIAL EFFECTS IN LOGISTIC MODELS


From   "Antonio Rodrigues Andres" <ara@sam.sdu.dk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: PARTIAL EFFECTS IN LOGISTIC MODELS
Date   Tue, 25 Nov 2003 16:59:09 +0100

Dear Stata users
I am estimating grouped data logistic models.  The problem with the
intepretation of these coefficients that they do not give true partial
effects.
My interest is to know how much an increase in income increase the
dep.variable.  The effect can be obtained from the partial derivative

g(x0+xb)*bj for variable xj. We need to evaluate the logistic density
function in the mean values of the variables. That's what I have done. I
got the same coefficient that when estimating the model. These
coefficients reported in STATA have straigthforward interpretation? how
i could get them?

glogit ntsex pop clsex $xvars $zvars $tvars
matrix b=e(b)
matrix list b
matrix accum a= clsex $xvars $zvars $tvars, mean(m)
matrix list m
matrix d=b*m'
matrix list d

global eff = ((exp(d))/(1+exp(d)))^2
di $eff *_b[income]

Regards 
Antonio
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index