Hello,
I am trying to calculate the mills ratio from the probit model on
participation (I do not need to do Heckman afterwards).
I tried to methods:
1- Heckman regression with the mills ratio specification
2- A solution proposed in the FAQ in stata that goes:
http://www.stata.com/support/faqs/stat/mills.html
1- probit model for participation
2- predict phat, xb
3- gen mills=exp(-.5*phat^2)/(sqrt(2*_pi)*normprob(phat) )
I am not getting the same results using either way. I would like to use
the second method
but it is not clear to me how this mills ratio is calculated given that as
I see it:
mills=f(xb)/F(xb) where f and F are density and cumulative functions of a
std normal
and that phat=F(xb)
thanks
*
* 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/
