st: Interval Regression and missing standard errors after using robust

[Lina C <linacs81@gmail.com>]

Dear all,
I'm running an interval regression using intreg. Without using the
robust option, the standard errors of all the variables are reported.
However, once I use the robust option some coefficients report missing
standard error (for one dummy variable. This is not collinearity
because the s.e. are reported without robust). I was wondering how can
I solve this or if I shouldn't use robust standard errors since
dropping the variable would change my initial results.  I have read
some past thread about it but in any of them the solution is offered.
I would be very grateful with any suggestion.

Many thanks,
Lina.

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