Dear all,

I have a binary dependent variable Y, endogenous variable X, control variables C1 and C2, and an instrument (binary) Z. My aim is to understand the impact of X on Y. I am using control function approach as follows:

regress X i.Z C1 C2
predict resid, residuals
probit Y X resid C1 C2

However, I realized that my instrument is also affecting C1. Could you please help me figure out whether/why it is a problem and please suggest any papers about this problem?

Thank you very much!