How can I create a multivariable Logistic regression model and use brms to make responses correlatted？

[zijun]

How can I create a multivariable Logistic regression model and use brms to make responses correlated？

For example, there are three alternatives for urban residents’ travel: car, bus, and subway. Therefore, I would like to use age, gender, and travel distance as independent variables, and region as grouping factor. How can I model these three alternatives as relevant?

the basic brms formular：
model= brm(choice ~ age+ gender+travel_distance+(1| region), data=data, family=‘categorical’)

I don’t know how to change this model to one with responses correlated.

[franzsf]

Brms does have some capability in conditional logit models through the nlf syntax. See:

Implement Conditional logistic regression models

[zijun]

Thanks！ It is very useful.