www.cmm.bristol.ac.uk/forum

Hi
I am trying to fit a two level ordinal model using runmlwin in stata with the following syntax, but I get this message " variable cons not found (error in option denominator())" . As I new with runmlwin would you please give me any suggestions.

runmlwin V_outcome5 cons V_SEXID V_job, level2( TheKey: cons ) level1( V_id:) discrete(dist(multinomial) link(ologit) denom(cons)) nopause


Best Wishes,
Ali

Can you confirm that you do have a variable called cons in your data?

ChrisCharlton wrote:Can you confirm that you do have a variable called cons in your data?

yes, but there is no variable with "cons" name.
I re-installed runmlwin again and try many times to fit the model, but I get the "mismatch" error! would you please help me.
Regards

If your constant isn't called cons you will need to provide its name instead. Alternative you can create a cons variable with the command: