David,
here's some sample code that might be helpful.
logit promoted african controls
// Adjust can do the next 4 lines at once, but
// I like the simple steps approach.
g aa = african
replace african = 0
predict eprom
replace african = aa
replace eprom = 0 if african==0
g actual = promoted * african
replace eprom = sum(eprom)
replace actual = sum(actual)
local shortfall = expected[_N]-actual[_N]
di "the shortfall is " %7.1f `shortfall'
// use the t-stat from the logit.
// Also you can put it in a program to clean up
// the display of the results you want.
> Hello.
>
> I'm running a promotion analysis using the following
> type of logit:
>
> promoted=african american + dummies + continuous vars
>
> Is there an easy way in STATA to calculate the
> shortfall in african american promotions (if one
> exists)?
>
> i.e. is there a shortcut to calculate and save the
> expected number of promotions?
>
> Thank you guys for any advice you may have
>
> -David
*
* 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/
