I posted some code for getting the C-stat a few days ago
http://www.stata.com/statalist/archive/2009-10/msg00267.html
I'm not familiar with -gologit-, but as long as you can produce
a predicted probability, the method should apply.
hth,
Jeph
[email protected] wrote:
Hi,
I have checked the archives and I can't find anything on this topic (sorry
if I have missed the answer to my question), but hopefully someone can
help me.
I have fitted a partial proportional odds model using -gologit2- in Stata
9.2 (3 levels of the outcome, 14 predictors). I used the -autofit- option
to assess proportionality and 2 of my predictors have non-proportional
odds.
Reviewers have suggested the use of a generalized c-statstic to assess the
internal validity of the model. I am struggling to work out how I might
get the value from Stata.
Please could someone give me a pointer in the right direction?
Best wishes,
Sara
*
* 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/
