ABSTRACT
In previous chapters, we have mostly focused on parametric mixed effects models. In these models, the random effects are typically assumed to follow (multivariate) normal distributions and the within individual (cluster) random errors are assumed to follow parametric distributions in the exponential family. For example, in LME and NLME models we assume that both the random effects and the responses (or the within individual errors) follow normal distributions, and in GLMMs we assume that the random effects follow normal distributions and the responses (or the within individual errors) follow parametric distributions in the exponential family. Likelihood inference is then based on the assumed distributions.
