Classification Table and Odds Ratios after mlogit

[Sarah Purnell]

I am running a multinomial logistic regression and I am trying to find additional statistics after using mlogit.

My DV has four outcomes and I have 3 continuous predictors and 2 categorical predictors.


-I am trying to create a classification table for my data (which I will then use to calculate the accuracy of my predictors). I see this is available for logit but not mlogit.

-I am also trying to figure out the odds ratios, but I have not been able to successfully do so for mlogit. I have found the risk ratios but am also looking for the odds ratios.

Any help would be appreciated!

[Maarten Buis]

You did not find risk ratios, but relative risk ratios. The relative risk ratios are the odds ratios, but StataCorp did not want to call them that. It is just a matter of choosing clear terminology, for which no good solution has been found. (see: https://www.stata.com/statalist/arch.../msg00085.html and the link therein) But that does not matter for you. Just use RRRs and know that they are odds ratios.

[Sarah Purnell]

Thank you so much for your response and the resource, Maarten! Coming from my discipline relative risk ratios and risk ratios are synonymous and I am just now discovering this is not the case for other disciplines.