Hi,
Consider the usual OLS regression in which two dummies reperesnent a
three-category categorical independent variable:
lny = a + b1x1 + b2x2 + b3x3 + e
where:
lny = log of dep var
x1 = dummy 1
x2 = dummy 2
x3 = continious control variable
In the ususal framework, the coefficients b1 and b2 are the differene in
percentages (roughly) between the categories in the regression and the
omitted category (i.e. the constant).
Instead, following Kennedy (1986), I want to get how all three categories
deviate from the overall mean of lny. That is, this procedure allows for
the estimation of all three dummies, and not just two.
Has anyone written a semi-canned procedure for this in Stata?
Thanks in advance.
Christer
Christer Thrane, dr.polit.
Associate Professor, Sociology
Department of Social Science
Lillehammer University College
2626 Lillehammer, Norway
+ 47 61 28 81 70 (fax)
+ 47 61 28 82 47 (phone, work)
+ 47 61 25 68 44 (phone, home)
+ 47 92 29 54 39 (mobile)
E-mail, work: [email protected]
E-mail, home: [email protected]
Kennedy, P. (1986): Interpreting Dummy Variables. The Review of Economics
and Statistics. 68, 174-175.
*
* 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/