Euler P. G. de Mello <[email protected]> asks about doing a constrained
regression with robust standard errors,
> After runing a constrained regression -cnsreg- and testing its
> validity, -whitetst- indicated that a robust estimation is needed. I
> would like to know how could such estimation be done using -cnsreg-?
There is no good reason why -cnsreg- does not support robust standard errors
(SEs), it is really just a matter of history. -cnsreg- was implemented long
before robust SEs were widely adopted in Stata and they were just not added.
Regardless, Euler can get robust standard errors from one of Stata's ML
estimators whose degenerate model is a linear regression -- say -arima- or
-intreg-. If Euler's -cnsreg- command is,
. cnsreg y x1 x2 x3 , constraint(...)
he can type,
. arima y x1 x2 x3 , constraint(...) robust
or,
. intreg y y x1 x2 x3 , constraint(...) robust
-- Vince
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
