re: st: loop

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

<>
Chiara said
I wrote a loop for wage regressions to obtain the predicted wages if
men are paid as men at each occupation:

reg lwage $x if fem==0
predict pip if fem==0, xb
forvalues k = 1 2 to 7 {
	predict pip`k' if fem==0 & occ`k'==1, xb
}
summarize pip, meanonly
scalar xMbM=r(mean) /*Predicted wages if men are paid as men*/
forvalues k = 1 2 to 7 {
	summarize pip`k', meanonly
	scalar xMbM`k'=r(mean) /*Predicted wages if men are paid as men at
each occupation*/
}

I now want to get the xMbM (and also xFbF for females) by occupation.
For each occupation I want the product between ythe mean individual
charachteristics (xM) and the coefficients (bM). I tried with this
loop, but i'm not sure it's the correct one:

reg lwage $x if fem==0
predict pip if occupation==1, xb
forvalues k = 1  {
	predict pip`k' if fem==0 & occ`k'==1, xb
}
summarize pip, meanonly
scalar xMbM=r(mean) /*Predicted wages if men are paid as men*/
forvalues k = 1 {
	summarize pip`k', meanonly
	scalar xMbM`k'=r(mean) /*Predicted wages if men are paid as men at
each occupation*/
}

More precisely, i'm not sure this wille give me each product between
mean values of characteristics (x, for M or F) and estomated wage
equation coefficients (b, for M or F) at each 1...7 occupation.




No need for all this manual labor...

--------------------
webuse nlsw88,clear
// lets treat marital status as equivalent of gender, since all people here are women
// run regression over all cases so can generate pred wage by mar.stat. and occup.
reg wage age collgrad south i.occupation
margins, by(married occupation) 
marginsplot,graph(married) xlab(,angle(90))
---------------------

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/