Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: defiing the reference category in multinominal logistic regression


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: defiing the reference category in multinominal logistic regression
Date   Sat, 28 Mar 2009 15:32:55 +0000 (GMT)

--- On Sat, 28/3/09, Dan Waldo <dan_waldo@yahoo.com> wrote:
> I am running Stata10/SE, and am doing a series of
> multinomial logistic regressions.
> 
> It's clear that Stata uses information from the
> char ageg[omit] 3 
> instruction (for example) when setting up the explanatory
> variables. But such is not the case for the dependent
> variable.
> 
> Again for example, the variable flushot is dichotemous with
> values 1,2 referring to yes,no.
> 
> -char flushot[omit] 2
> -xi: svy: mlogit flushot i.adjsmoker i.ageg i.rostsex
> i.myrace i.my_income i.my_ed i.my_dwell i.placepar , rr
> 
> produces a regression in which (flushot==Yes is the base
> outcome)
> 
> -xi: svy: mlogit flushot i.adjsmoker i.ageg i.rostsex
> i.myrace i.my_income i.my_ed i.my_dwell i.placepar , rr b(2)
> 
> produces a regression in which (flushot==No is the base
> outcome)

If you have a dichotomous depedent variable you are better of 
using -logit- rather than -mlogit-. Both are legal, but 
-logit- is easier.

If you still want to use -mlogit- you can specify the base 
outcome with the -baseoutcome()- option. see: -help mlogit-.

Hope this helps,
Maarten

-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://home.fsw.vu.nl/m.buis/
-----------------------------------------





      

*
*   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/



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