st: clear estimated coefficient

[natasha agarwal <agarwana2@googlemail.com>]

Dear All,

Please accept my apologies for the previous post.

I want to estimate production function for each industry and calculate
the predicted output for each industry. To do so, I estimate the
following regressions

levpet lnrval if inm==1, free(lnw) proxy(lnb) capital(lnk) valueadd

gen yhat1 = _b[lnk] * lnk +_b[lnw]* lnw if inm==1


levpet lnrval if inm==2, free(lnw) proxy(lnb) capital(lnk) valueadd

gen yhat2 = _b[lnk] * lnk +_b[lnw]* lnw if inm==2


however yhat2 takes _b of the regression where inm==1 and not inm==2.

I tried using clear matrix, scalar drop _all, but the results do not change.

Can you please help me with the same ?

Please note that levpet is a user written programme available as a
stata module by levinsohn and petrin in their co-authored paper in
2003. Stata Journal article of the same can be accessed at
http://www.stata-journal.com/sjpdf.html?articlenum=st0060. The same
can be downloaded as ssc install levpet

Thanks,
Natasha
*
*   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/