Mixed Models

Objectives

Specific features of hierarchical (longitudinal) data.
Historical methods for analyzing grouped data (ANOVA for repeated data, etc.)
Linear random effects model (Definition)
Inference by maximum likelihood or residual likelihood
Best linear unbiased prediction (BLUP), Model fit, Interpretation).
Prediction of random effects.
Generalized linear mixed model.

Course outline

Differentiate between fixed and random effects in a mixed model.
Explain the maximum likelihood principle and the REML principle.
Implement a mixed model using statistical software (your choice of R or SAS).
Check the goodness-of-fit of the mixed model.
Interpret the results of mixed-model inference.
Identify the use of mixed models in epidemiology and clinical research.
Describe mixed generalized linear model inference.

Prerequisites

Probability, inferential statistics, SAS, R (1A)
Linear regression, GLM (2A)