For fitstat, the advice given in post #4 still holds.
The fact that your dependent variable is ordinal has nothing to do with the errors you are getting when using omodel. If a command says it does not support factor variables, it means you have to generate the indicator variables for the independent variables yourself. If the variable is already coded 0/1 you don't have to do anything other than drop the i. notation. If it is a categorical variable with more than 2 categories, do something like
tab religion, gen(relig)
If you have further Qs, Nick's advice to start a new thread is a good way to go. Otherwise people who know the answers might not even look at your post because the thread's title refers to Stata 11 which they might not know anything about.
I might add useful information to an old thread if I think it could help someone who finds that thread in the future. But I rarely add new Qs because they can easily get overlooked. Better to start a new thread, perhaps including a link to the old one.
The fact that your dependent variable is ordinal has nothing to do with the errors you are getting when using omodel. If a command says it does not support factor variables, it means you have to generate the indicator variables for the independent variables yourself. If the variable is already coded 0/1 you don't have to do anything other than drop the i. notation. If it is a categorical variable with more than 2 categories, do something like
tab religion, gen(relig)
If you have further Qs, Nick's advice to start a new thread is a good way to go. Otherwise people who know the answers might not even look at your post because the thread's title refers to Stata 11 which they might not know anything about.
I might add useful information to an old thread if I think it could help someone who finds that thread in the future. But I rarely add new Qs because they can easily get overlooked. Better to start a new thread, perhaps including a link to the old one.
Comment