I am trying to estimate an ordered probit model, where the data in question comes from two separate counties. I was thinking that I should use fixed effects to account for unobserved heterogeneity across the counties, but apparently a fixed effects ordered probit model is not possible to estimate. I read somewhere else that you can approximate the fixed effects model by estimating a random effects model AND including a dummy for the variable in question.

Am I on the right path with my thinking here? Is it possible to estimate a random effects ordered probit model with 2 different outcomes for the random variable? Or am I ok to estimate a regular ordered probit model and simply include an indicator variable for "county"?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.