The outcome is ordinal 1 to 10 and predictors 1 to 3, no zero value and no missingness.
I am running ordinal regression on multiple ordinal predictors.
The model is:
prior_bodn <- prior(normal(10, 100), class = "b") + prior(normal(1.1, 100), class = "Intercept") fit_bodn<-brm( formula = bodn ~ 1 +mo( hid)+mo( ch)+mo(ht)+time+(1 |id), data = d, family = cumulative("probit"),prior_bodn = prior )
I am getting this message within the analyses
Rejecting initial value:
Chain 1: Log probability evaluates to log(0), i.e. negative infinity.
Chain 1: Stan can’t start sampling from this initial value.
I am quite new to brms. Is this message OK? Should I be concerned? Thanks