I'm using probit to calculate IMR also, Heckman command is not applicable in my model.
But why do I have different formula for IMR calculation?
mine is:
IMR=normden(-phat)/(1-norm(-phat))
-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of ZYD
Sent: Saturday, March 12, 2005 8:51 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: Why estimation in first step of two-step Heckman model is
different from Probit model for selection
I estimated two-step Heckman model using
heckman y x1 x2, select(z1 z2) two
predict mill1, mill
I also estimated Probit model using
probit y z1 z2
predict phat, xb
gen mill2=exp(-.5*phat^2)/(sqrt(2*_pi)*normprob(phat))
why the coefficients for z1 z2 in two estimation are different in the
estimating procedure? I also calculate the inverse mill's ratio (IMR)
for the probit model, and not surpriingly, the two IMRs are different?
Which one is more reliable when y is a binary variable? Thanks!
sincerely
yiduo
*
* 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/
--------------------
This email message is for the sole use of the intended recipient(s) and
may contain privileged information. Any unauthorized review, use,
disclosure or distribution is prohibited. If you are not the intended
recipient, please contact the sender by reply email and destroy all copies
of the original message.
*
* 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/