I find the original question unclear.
My take is the following: -ologit- is the proportional
odds versions of ordinal logistic regression (there are
others). In this version, the various coefficients
refer to any comparison (that is why it is called
proportional odds -- the odds do not change across
comparisons. The cutpoints (Stata terminology) refer
to the categorization into your three groups; this is
discussed on p. 98 of the version 8 manual under -oligit-.
Rich
[email protected] wrote:
>
> hm, I don't think ordered regression has a reference group, it just gives
> you one b which estimates the effect of X over all the outcomes, assuming
> proportional odds. There is a reference group in multinomial logit (mlogit)
> which does not assume an ordering (e.g., professions as outcome categories).
> You can change the reference group in multinomial regression using the
> "basecategory(whatever)" option.
>
> Henry
>
> -----Urspr�ngliche Nachricht-----
> Von: [email protected]
> [mailto:[email protected]]Im Auftrag von Abomohammad
> Q. Kamaluddin
> Gesendet: Sonntag, 12. Oktober 2003 21:22
> An: [email protected]
> Betreff: st: ologit
>
> Hi list member,
> I�m using the ologit model that has three categories.
> The output I get refers to the last group (group that
> takes the value of 3) and gives result accordinly.
> My question is, how would I specify that the output
> should take group 1 or group 2 as the reference group?
>
> Thanks
> Q
>
> http://search.yahoo.com.au - Yahoo! Search
> - Looking for more? Try the new Yahoo! Search
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
