Hi Fred,
Because you used the pr option (or the default after logit) you were in
effect applying the logit transformation to the linear predictor. With
-predict- you are first applying the transformation and then compute
the mean, while with -adjust- your are first computing the mean and
than apply the transformation. These are not the same for non-linear
transformations, like the logistic transformation. You can make the
results from -predict- and -adjust- the same by not applying the
transformation, i.e. ask for the linear predictor xb, and this is what
I tried to show with the example.
Hope this is clearer,
Maarten
--- Fred Wolfe <[email protected]> wrote:
> Hi Maarten,
>
> I did not use predict, xb. I used predict p, in which the default is
> -pr-. You cited the OLS command I showed above, but not the predict
> command I used for LR.
>
>
> Using logistic regression with -pr- prediction and adjust with -pr-,
> I obtained the following results:
>
> From tab 189 .3121693 .4646093
> From predict,pr 189 .3121693 .1481592
> From adjust,pr 189 .289227
>
> Fred
>
>
> At 08:00 AM 12/23/2006, you wrote:
> >--- Fred Wolfe <[email protected]> wrote:
> > > The manual and the FAQ state that the results after predict and
> > > adjust are the same. <snip> However, when I use the logistic
> > > regression example from page 20 (Reference A-J) for the adjust
> > > command, predict and adjust give different results.
> >
> >The key is that by looking at predicted probabilities you apply the
> >logit transformation to the linear predictor. If you don't apply the
> >transformation than the results will be the same again:
> >
> >*-------------- begin example -------------
> >use http://www.stata-press.com/data/r9/lbw,clear
> >xi: logistic low age lwt i.race smoke,nolog
> >predict p, xb
> >su p
> >adjust
> >*--------------- end example --------------
> >
> >Hope this helps,
> >Maarten
> >
> >
> >
> >-----------------------------------------
> >Maarten L. Buis
> >Department of Social Research Methodology
> >Vrije Universiteit Amsterdam
> >Boelelaan 1081
> >1081 HV Amsterdam
> >The Netherlands
> >
> >visiting address:
> >Buitenveldertselaan 3 (Metropolitan), room Z434
> >
> >+31 20 5986715
> >
> >http://home.fsw.vu.nl/m.buis/
> >-----------------------------------------
> >
> >Send instant messages to your online friends
> http://uk.messenger.yahoo.com
> >*
> >* 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/
>
>
> Fred Wolfe
> National Data Bank for Rheumatic Diseases
> Wichita, Kansas
> Tel +1 316 263 2125
> [email protected]
>
>
> *
> * 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/
>
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
Inbox full of spam? Get leading spam protection and 1GB storage with All New Yahoo! Mail. http://uk.docs.yahoo.com/nowyoucan.html
*
* 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/
