On parameters estimation procedures in multilevel models

We examine maximum likelihood estimation procedures in multilevel models re-latedto two-level hierarchically structured data. Usually, for fixed effects and variance com-ponentsestimation, multivariate normal distribution is assumed. Here we consider for randomeffects multivariate exponential power distribution (MEP), which represents one of the possiblegeneralizations of multivariate normal distribution. We examine robustness of maximum like-lihoodestimators under normal assumption when, indeed, random effects are MEP distributed.The study is conducted through MC simulation procedures