st: initial value not feasible with -uhet- option in frontier command

[Yi CHEN <Yi.Chen@u-clermont1.fr>]

Hi,

I read  David (ddrukker@stata.com)'s answer to a question concerning the 
"intial values not feasible" message with command "frontier".
My question is, if I include a -uhet- option in the command, and got the 
"initial values not feasible" error message, how can I correct the problem? 
If I do not put the variables in -uhet- in OLS regression, I will get the 
error message as "initial vector: matrix must be dimension ** (the number 
of variables after frontier + the number of variables in -uhet- option)". 
But will there be problem if I add simply the variables in -uhet- in the 
OLS regression?
Hope I've made clear the question.
Many thanks.

Chen 
*
*   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/