Assumptions ordered probit model

[Eisenn]

Hi,

I am running an ordered probit regression, and I was wondering if there are any assumptions my data needs to fullfill in order to fully rely on the results. And if any assumptions need to be met how can I test for them and if needed, how can I fix them?

[Carlo Lazzaro]

Eisenn:
your question is probably too vague for getting helpful replies.
I would recommend you to start out with -help oprobit- and related entry in Stata 13.1 .pdf manual.
As an aside, please note the preference for real full names on ths forum (as per FAQ #6).

[Richard Williams]

You can start with the parallel regressions assumption, which can be tested with the user-written routines -omodel- and -gologit2-. See their help files for more.