st: Ordered probit with endogenous regressor

[Dario Maimone Ansaldo Patti <dmaimone@unime.it>]

Dear Stata Users, 
  
I am estimating an ordered probit where one of the covariates is endogenous (it could be either a binary dummy or an ordered variable). Therefore I set up a first equation to “endogeneize” my regressor in the second equation. I use cmp to estimate my model. 
  
My question is the following: 
  
I hown that atanhrho is zero when rho is zero. When I look at the p-value associated to atanhrho, I can reject the null that the parameter is equal to zero. This is an evidence that the regressor is, in fact, endogenous. Am I correct? Could I conclude that the choice of instruments I have done in the first equation is valid?   Thanks for your attention.   

Dario Maimone Department of Economics University of Messina 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/