I am working with a binary outcome model so yi takes the value 1 or 0. Xi is a vector of 8 explanatory variables and Beta (B) is a vector of 9 parameters (8 variables +intercept).

I am working with the probit model so F(X'B)=standard normal c.d.f.

Next;

Let y(hat)i be equal to 1 if the standard normal cdf evaluated at XiB > 0.5

Let y(hat)i be equal to 0 if the standard normal cdf evaluated at xiB <=0.5

I have two questions:

1. For how many n does yi=y(hat)i ?

My understanding:

To do this I need to use stata to see for how many n does yi=1 and then for how many of these does the standard normal c.d.f evaluated at XiB>0.5. Then do the same for yi=0.

2. For how many yi=y(hat)i=1 ?

I dont understand this question as well. Is it not part of the first question?

Any help would be greatly appreciated.