Fit a linear mixed effects model for censored data with Student-t or normal distributions. The errors are assumed independent and identically distributed.