 Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: conditional probabilities after biprobit

 From george joseph To statalist@hsphsun2.harvard.edu Subject st: conditional probabilities after biprobit Date Mon, 31 Jan 2011 11:09:19 -0500

```after biprobit.

I have the following code

biprobit (y1 = y2 xvars) (y2 z xvariables ), robust

After mfx, compute, I try to calculate the conditional probability
of-  y1=1 conditional on y2=1 and then  y1=1 conditional on y2=0. I
found > the following faq on the stata list useful.

http://www.stata.com/statalist/archive/2010-/msg00078.html
The following suggestion is provided. However, I am not sure whether
this calcualation is indeed correctt, because it does not take into
consideration the complementary events. Also when we make all x equal
to 1 and then all x equal to 0, what exactly are we trying to do here.

sysuse nlsw88, clear
g x=race==1
g y=married
biprobit (y=x grade south smsa) (x=industry occupation union)
mfx
margins, dydx(*)
g wasx=x
replace x=1
predict p1a, p11
predict p1b, p10
predict p1c, p01
predict p1d, p00
g p1=p1a/(p1a+p1c)
replace x=0
predict p0a, p11
predict p0b, p10
predict p0c, p01
predict p0d, p00
g p0=p0b/(p0b+p0d)
replace x=wasx
drop wasx
g dp=p1-p0
su dp

Could anyone in the list clarify thiss

Thanks.
George

*
*   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/
```