Extending Multivariate-t Semiparametric Mixed Models for Longitudinal data with Censored Responses and Heavy Tails

In this paper we extended the semiparametric mixed model for longitudinal censored data with
normal errors to Student-t erros. This models allows exible functional dependence of an outcome
variable on covariates by using nonparametric regression, while accounting for correlation between
observations by using random eects. Penalized likelihood equations are applied to derive the
maximum likelihood estimates which appear to be robust against outlying observations in the
sense of the Mahalanobis distance. We estimate nonparametric functions by using smoothing
splines jointly estimate smoothing parameter by the EM algorithm. Finally, the performance of
the proposed approach is evaluated through extensive simulation studies as well as application to
dataset from AIDS study.