Statalist


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

Re: st: After mlogit: Graphing predicted probabilities for transition between 2 subgroups


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: After mlogit: Graphing predicted probabilities for transition between 2 subgroups
Date   Mon, 19 Oct 2009 05:11:26 -0700 (PDT)

--- On Mon, 19/10/09, Stefan Baron wrote:
> I incorporated some interaction effects into mlogit models
> and now I  want to illustrate the additional effect graphically.
> I tried prgen, but it computes the prediced propabilities
> for ending up in the outcome category in relation to all other
> categories. Now, I wonder whether there is a tool for computing the
> predicting probabilities for ending up in one category in relation to
> ANOTHER category, not all categories?

exp(linear predictor) will give you the ratio of the probability of
entering the group belonging to that linear predictor over the 
probability of entering the refernce categorie. One way to get those
is to use -predictnl-, as is shown in the example below:

*--------------- begin example ------------------
sysuse auto, clear
recode rep78 1/2=3
mlogit rep78 mpg
predictnl pr4overpr3 = exp(xb(#1))
predictnl pr5overpr3 = exp(xb(#2))
twoway line pr4overpr3 pr5overpr3 mpg,  sort ///
   legend(order(1 "good versus average"      ///
                2 "excelent versus average") ) 
*----------------- end example -------------------
(For more on examples I sent to the Statalist see:
http://www.maartenbuis.nl/example_faq )

Hope this helps,
Maarten

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

http://www.maartenbuis.nl
--------------------------


      

*
*   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