*
* Example 7.6 on page 227
*
open data wage1.raw
data(format=free,org=columns) 1 526 wage educ exper tenure nonwhite female married numdep smsa northcen $
  south west construc ndurman trcommpu trade services profserv profocc clerocc servocc lwage expersq tenursq
*
* These show how to create the squared series
*
set expersq = exper**2
set tenursq = tenure**2
*
* And these are the interactive dummies
*
set marrmale = married*(1-female)
set marrfem  = married*female
set singfem  = (1-married)*female
set singmale = (1-married)*(1-female)
*
linreg lwage
# constant marrmale marrfem singfem educ exper expersq tenure tenursq
*
* Using married women as the base group
*
linreg lwage
# constant marrmale singmale singfem educ exper expersq tenure tenursq
*
* An alternative is to redoing the regression is to use the SUMMARIZE
* instruction, which computes the value and standard error of linear combinations
* of the coefficients of the last regression. For instance, the following will
* give the estimate of the difference between married and single males, which
* could be read off directly from the first form but not the second.
*
summarize(vector=||0,1,-1,0,0,0,0,0,0||)