[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Restated: shortfall analysis - logit question
Thank you Stephen. I will be more specific. From the logit model I
would like to be able to determine the overall shortfall of
promotions by calculating the expected number of promotions
(predicted from the model) for the minority group and subtracte this
from the actual number of promotions. Is there a nice Stata
shortcut that does this and also gives a t-Stat for this shortfall?
Thanks everyone (again)
--- In email@example.com, "Stephen P Jenkins"
> > -----Original Message-----
> > From: owner-statalist@h...
> > [mailto:owner-statalist@h...] On Behalf Of
> > David J. Bernstein
> > Sent: 01 June 2004 15:57
> > To: statalist@h...
> > Subject: st: shortfall analysis - logit question
> > 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)?
> You need a definition of "Shortfall".
> From your model, you can calculate predicted probabilities of
> for african-americans and predicted probabilities of promotions for
> others, holding the other covariates at fixed values chosen by
> Have a look at -adjust-, -predict-, and related Manual entries.
> For decomposition of a 0/1 vble [analogous to Blinder/Oxaca
> decompositions of a continuous vble] based on estimation of
> equations for african-americans and for others, see Gomulka J &
> (1990), 'The employment of married women in the United Kingdom,
> 1970-83', Economica, 57, 171-199.
> Professor Stephen P. Jenkins <stephenj@e...>
> Institute for Social and Economic Research
> University of Essex, Colchester CO4 3SQ, U.K.
> Tel: +44 1206 873374. Fax: +44 1206 873151.
> * 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/
* For searches and help try: