st: svyintreg vs. intreg with pweights

["Sergiy Radyakin" <Radyakin@aoek.uni-hannover.de>]

Hello Statalisters,

I have survey data on incomes (in brackets) and pweights are provided as 
well.
I would like to find the average income. To do this, I regress the interval 
ends on a constant using -intreg-

This command works:

   intreg inc1 inc2 [pw=my_weight_variable]

This syntax does not work:

   version 8
   svyset [pw=my_weight_variable]
   svyintreg inc1 inc2

(fails after -svyintreg- with the error message "initial values invalid").
There are no zero or missing weights (if it is of any importance).

My questions are:

1) are the above two syntaxes equivalent?
2) if the answer to #1 is negative, then what is the rationale behind 
choosing one vs the other?
3) how can I specify that -svyintreg- takes the estimates from 
unweighted -intreg- as initial values?
  Or more generally, how to make the svy version to work in my case?

Thanks,
  Sergiy 
*
*   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/