Statistical models of young stellar clusters enable us to compare model predictions to observations while incorporating the particularities of the data, like heteroscedastic uncertainties, missing values, zero point callibrations, and a variety of correlations.
I will present two Bayesian hierarchial models that were designed to infer diverse properties of young stellar clusters. One of them takes dataset of hundreds of thousands of sources in a possible highly extincted sky region and simultaneously identify both cluster members and the cluster luminosity distribution. The second model is desinged to simultaneously infer the 3D structure of a stellar cluster and the individual positions of it stars. This model has been tailored to fit the Gaia data and deals with the uncertainties and the parallax spatial correlations.