At 11:38 PM 10/6/2009, Emmanuel Guindon wrote:
If I am not mistaken, gologit and gllamm with -thres- option can both
be used to relax the parallel regression (or proportional odds)
assumption inherent in ordinal logits (ologit). gologit and gllamm
with -thres- option produce results that are more or less equivalent
but of opposing signs. Do coefficients obtained with gllamm -thres-
have the same interpretation as the ones obtained using gologit?
-Emmanuel
I don't see why not. Having said that, there are different ways to
interpret the results of a gologit model. You may want to look at
http://www.nd.edu/~rwilliam/gologit2/RWNASUG2006.pdf
http://www.nd.edu/~rwilliam/gologit2/RWNASUG2006_Handout.pdf
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME: (574)289-5227
EMAIL: [email protected]
WWW: http://www.nd.edu/~rwilliam
*
* 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/
