re: Re: st: Constructing a variable from standard deviations

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

<>
M.P. said

Do I have to run the regression seperately for all occupations or can 
run a
regression with all the dummies?

It seems to me that you use a variable with different categories 
instead of
individual dummies. Do you think I should also create such a variable 
for my
106 studyfields?

What other adjustments are nessacary  to use your example for my 
modelling
purposes?


I do not see why this is not a straightforward application of -collapse- per the example I posted. There should
be no need to resort to maximum likelihood techniques. I agree with Maarten: you should estimate the model
separately for the 106 different groups, per my example from grunfeld.dta.

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/