st: Equal dummy variable coefficients across equations

[Chun Li <c.li29@student.unimelb.edu.au>]

Dear Statalist,

I need to estimate a system of equations in this form:

y1 = par1 + par2 +  fe_a1 + fe_a2 + fe_a3
y2 = par1 + par2 +  fe_a1 + fe_a2 + fe_a3
...
yn = par1 + par2 +  fe_a1 + fe_a2 + fe_a3
etc.

Par1 and par2 are slope parameters, fe_a1, fe_a2 and fe_a3 are fixed
effects parameters.  Each observation takes on one of the fixed effect
values (i.e. the constant term replaces one of them when estimating).

I want to estimate these equations by simple linear OLS, but **I
cannot use sureg, because**
a) Coefficients for Par1 and Par 2 needs to freely take on
**different** values across equations.
b) Fixed effects coefficients fe_a1, fe_a2 and fe_a3 need to contain
the **same** value across equations.

Please kindly advise on how to do this!

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