How to handle large number of observations below LLOQ

[jroon]

@daszlosek you might find this similar thread interesting which threw up a few options:

Modelling outcome data with a limit of detection. Is a hurdle model appropriate?

Ah ok. I did some tests of ordinal models in Stan and was having problems all right. I’m no really fmailiar with rstanarm but perhaps I will take a look. Thanks Frank.