<>
You probably want -normal()- there....
*************
// to replicate millsratio for twostep heckman
cap ssc inst indeplist
cap ssc inst tomata
webuse womenwk, clear
g wageseen=!missing(wage)
probit wageseen married children educ age
indeplist
tomata `r(X)'
mata
b=st_matrix("e(b)")
Z=(married, children, education, age, J(2000,1,1))
millsratio=normalden(Z*b'):/normal(Z*b')
st_addvar("double","matamills")
st_store((1,2000),"matamills",millsratio)
end
heckman wage educ age, select(married children educ age) first ns(mills)
nolog two
compare mata mills
*************
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von ??
Gesendet: Mittwoch, 6. Mai 2009 10:44
An: statalist
Betreff: RE: Re: st: AW: Question about Heckman Test
Dear STATA List!
When I did the probit in Heckman test, the "norm" function did not work in
my STATA.
Can someone let me know how the norm function as shown below will work?
probit s age student educat goodhealth
predict p1, xb
gen phi=(1/sqrt(2*_pi))*exp(-(p1^2/2))
gen capphi=norm(p1)
gen invmills=phi/capphi
Thank you!
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
