Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Logistic regression interpretation


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Logistic regression interpretation
Date   Thu, 23 Sep 2010 07:31:52 +0000 (GMT)

--- On Wed, 22/9/10, Dr. Bill Westman wrote:
> after the margins command, we have the four outcomes. 
> The last line (1:1) in your example shows:
> 
>  1 1  |   .4279095 
> _b[1.married#1.collgrad] in my example  17.87322.
> How do you interpret the 17.87 margin?

Given the way I used the -margins- command (with the 
-exp(exp(xb()))- option), it will produce odds. This is a 
measure of the likelihood of death similar to but not the
same as a probabilty. The 100*probability gives you the 
expected number of people that died per 100 persons, while
the odds gives you the expected number of people that died
per person that survives. So 17.87 means that you expect
to find within the high risk treatment group 17.87 dead 
people for every living person. (Assuming that your dependent
variable is coded 1 "died", 0 "survived".)
 
> Also - those two di statements at show the exact same odds
> - was this intentional?

Yes, odds ratios are literaly that: ratios of odds. My example
was intended to show that, and that this odds ratio is constant
regardless of whether one is in the high risk or low risk group.

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   |   Site index