re: Re: st: playing with coefficients

[Christopher Baum <kit.baum@bc.edu>]

<>
Sure, I did not state the meaning of playing with coefs from different
models. The reg are wage equations by gender and occupations. All
those info will be needed to compute an overall wage decomposition
across (obviously) gender and all the occupational cats.

Anyway, statsby seems to be the solutio. I'm going more in detail


Not at all clear why you want to run separate regressions for gender and occupation categories. There are ways to cast this as a single regression, which would surely make things simpler in terms of comparisons, and would enable formal tests for differences---which you will not have with separate equations.

Kit


Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                             An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
  An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   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/