Distribution-Free Estimation of Heteroskedastic Binary Response Models in Stata

Jason R. Blevins and Shakeeb Khan

• Working Paper: dfbr.pdf (second version, 2009–01–09)
• BibTeX record: dfbr.bib
• Stata program: dfbr.ado (version 0.8.1)
• Stata help file: dfbr.sthlp

Both the Stata program and the paper are works in progress. Comments are welcome.

Abstract. This paper considers two recently developed semiparametric estimators for distribution-free binary response models exhibiting multiplicative heteroskedasticity under a conditional median restriction. It shows that these estimators can be implemented in Stata through simple modifications to the nonlinear least squares Probit criterion function. Both estimators are implemented in a new Stata command, dfbr, and several examples of its usage are provided.

Keywords: dfbr, binary response, heteroskedasticity, Probit, semiparametric estimation, sieve estimation.

JEL Classification: C13, C14, C25, C87.

History:

• September 6, 2008: first revision of working paper; dfbr version 0.8.0.
• January 9, 2009: second revision of working paper; dfbr version 0.8.1.