This seminar offers an introduction to multilevel modelling in its frequentist and Bayesian flavors. The course consists of three parts: part one reviews linear and generalized linear models, and introduces basic concepts of multilevel modelling such as nested and nonnested data structures, fixed and random effects, and variance components. The second part covers the principles of Likelihood and Bayesian inference, and demonstrates how multilevel models are fitted using R and Stan statistical computing environments. Part three offers the students the opportunity to apply those methods in their own research, and to discuss problems and results.
Students will be familiarized with concepts of multilevel analysis and alternative methods of statistical inference used for fitting multilevel models. Students will also learn how to use statistical software to fit multilevel models to data, and to interpret statistical results.
The course takes place on Tuesdays 11.45-13.15 in C-421.