Mixed Models: Theory and Applications

Mixed models can be used for the following purposes: • To model complex clustered or longitudinal data. • To model data with multiple sources of variation. • To model biological variety and heterogeneity. • As a compromise between the frequent ist and Bayesian approaches. • As a statistical model for the penalized log-likelihood. • To provide a theoretical basis for the Healthy Akaike Information Criterion (HAIC). • To cope with parameter multidimensionality. • As a statistical model to solve ill-posed problems, including image reconstruction problems. • To model shapes and images.