I am away from the 21st September until the 12th October. Please direct any urgent enquiries to Linda Vesel (linda.vesel@lshtm.ac.uk).
Seyi
>>> Maarten Buis <maartenlbuis@gmail.com> 09/22/11 08:38 >>>
On Thu, Sep 22, 2011 at 12:10 AM, Roger Cremades Rodeja wrote:
> If I present the results of the proportional Logit in odds ratios, the odds
> ratio is related to having a proportion of 1?
> say that for a dummy indep var the odds ratio is 4, does it mean that when
> dummy=1 "an outcome value of proportion equal to 1" is 4 times more likely
> than when dummy=0, do you know if it has been used in any other paper?
If you specify the -eform- option after a fractional logit model
(-glm- with the -link(logit)- and -vce(robust)- options) than you will
not get odds ratios but what I have called relative proportion ratios.
There is a discussion on how to use these in the help-file of
-betafit-. That discussion is also directly applicable to relative
proportion ratios from a fractional logit model.
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/
*
* 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/
