Hierarchical Multinomial Model Optimization/Specification

Modeling
, ,
15

The main source of your divergences is probably this part:

  target += multinomial_logit_lpmf(Coded[N] | [0, muBuilding[N], muCaregiver[N], muPatient[N],
                                                   muPatientpsych[N], muSDOH[N], muCanmeds[N],
                                                   muExpertise[N], muRecommendations[N]]');
    }

Because N is just the total number of individuals, you’re only specifying the multinomial_logit_lpmf prior for the very last individual. You’ll need to have this as a loop (which is what I think it used to be?) and that should be a bit nicer with the divergences

1 Like