Choosing the right likelihood for your data

[sammosummo]

Sounds like censored regression (not to be confused with truncated regression!). There’s been quite a bit of discussion about these models here and now there is a tutorial in the docs.