It is easy to setup a model with a single Nuisance Q block that is applied to multiple relative indices.
Casal2 should report an error for this situation. Each time an observation is calculated in the likelihood it will generate a new nuisance q for that particular observation. The nuisance q reported will be for the last observation that asked it to be calculated making it look like there was only a single q when instead there were multiple calculated for that model run.
It is easy to setup a model with a single Nuisance Q block that is applied to multiple relative indices.
Casal2 should report an error for this situation. Each time an observation is calculated in the likelihood it will generate a new nuisance q for that particular observation. The nuisance q reported will be for the last observation that asked it to be calculated making it look like there was only a single q when instead there were multiple calculated for that model run.