The new meintreg
command fits models in which the outcome is interval measured (interval-censored) and the observations are clustered.
Interval measured means that rather than the outcome (y) being observed precisely, it is known only that yl ≤ y ≤ yu in some or all observations. Observations can also be left-censored (y ≤ yl) or right-censored (y ≥ yu). An interval-measured outcome might be income, recorded in income brackets, or weekly minutes of exercise, recorded as less than 30 minutes, 31-59 minutes, 60-89 minutes, etc. Multilevel mixed effects means that the fitted model accounts for clustering, such as when people live near each other or students attend the same school or students are tested repeatedly.