Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: logistic, dummies: some observations not used, why, and how to change?


From   "annelieswerk" <a.vos@erasmusmc.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: logistic, dummies: some observations not used, why, and how to change?
Date   Mon, 9 May 2005 20:30:43 +0200

somehow my message didn't get through, sorry if you get it double now...

Dear all,
I think this should be anywhere in the manuals, but I can't find it,
probably since I don't know where to search, so maybe you can help?
I use Stata 8.0 SE.
I want to do a logistic regression, with outcome variable CLPL (which can be
0 or 1), and predictor variables EXPOSE (0 or 1) and ORIGIN (10 possible
values, so I created dummies, named _DORnameofcountryorregion).

Here is some output:
----- 
. logistic CLPL EXPOSE
Logistic regression   Number of obs   = 299
   LR chi2(1)      = 1.02
   Prob > chi2     = 0.3118
Log likelihood = -147.99309   Pseudo R2       = 0.0034

. logistic CLPL _DOR*

note: _DOR_latamerica != 0 predicts failure perfectly
      _DOR_latamerica dropped and 8 obs not used

Logistic regression                               Number of obs   = 291
                                                  LR chi2(5)      = 9.35
                                                  Prob > chi2     = 0.0958
Log likelihood = -142.04343                       Pseudo R2       = 0.0319
---- 

My question is: yes indeed, all 8 records on latin america do have CLPL==0,
but I do want to include those observations when I estimate the effect of
ORIGIN on my outcome (as such or in combination with EXPOSE or other
variables). Why does Stata exclude these observations, and what should I do
different?

Thanks for any advice,
Annelies

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index