. gen y= event/trial

. gen x=ln( dose)
(1 missing value generated)

. probit y x

outcome does not vary; remember:
                                  0 = negative outcome,
        all other nonmissing values = positive outcome
r(2000);

Since no observation has a y-vlaue of zero, probit model cannot be built here.

clear

input dose event trial
0.0 3  15
1.1 4  15
1.3 4  15
2.0 3  15
2.2 5  15
2.8 4  15
3.7 5  15
3.9 9  15
4.4 8  15
4.8 11 15
5.9 12 15
6.8 13 15
end

bprobit  event trial dose
gprobit  event trial dose

blogit  event trial dose
blogit  event trial dose,or

glogit  event trial dose
glogit  event trial dose,or