Below is a link to a program I wrote to conduct tests of stochastic dominance using Stata. It also takes advantage of Stata’s ability to handle complex sampling design in the estimation of the standard errors.
If you have any comments on the program (negative criticism is especially welcome), please send an e-mail to me at firstname.lastname@example.org. Thanks!
The World Bank also has a very useful Poverty Analysis Toolkit that has downloadable Stata programs to do (a) growth-inequality decompositions, (b) sectoral decompositions of inequality, (c) growth incidence curves, and (d) stochastic dominance analysis. It also has programs to estimate limited dependent variable models with endogenous explanatory variables (IV probit), and endogenous switching regression models.