Hello,
I am estimating the balanced panel data for the model where the dependent variable is a fraction. Both endogenous and instrumental-variable are binary (0/1). I have read the paper "Panel data methods for fractional response variables with an application to test pass rates" in which the authors considered the endogenous variable as continuous and modeled accordingly. However, in my case, it's not the same. Therefore, I kindly request if anyone could give advice on how to express the conditional mean model presented in the aforementioned paper (equation 4.1) and how should I estimate the residuals from first stage regression?

Thank you very much in advance.

Kind regards,
Faruque.