Distribution-Free Estimation of Heteroskedastic Binary Response Models in Stata

This paper considers two recently proposed semiparametric estimators for distribution-free binary response models under a conditional median restriction. It shows that these estimators can be implemented in Stata using nl through simple modifications to the nonlinear least squares probit criterion function. We then introduce dfbr, a new Stata command which implements these estimators, and provide several examples of its usage. Although it is straightforward to carry out the estimation using nl, the dfbr implementation uses Mata for improved performance and robustness.