Description: Mixed models provide a flexible framework for analyzing data with multiple sources of random variation and they are indispensable in many medical, biological, and engineering applications. When treatments are tested in medical applications, the responses for individuals receiving the same treatment often vary due to unobserved genetic factors and this variation must be taken into account when comparing ifferent treatments. Similarly, in agricultural field trials, random soil variation affects the yield within plots. In quality control applications, the variability of the output of a production process may, apart from random noise, e.g. depend on the batches of raw material used and the employee involved in the manufacturing process. The course will provide an introduction to statistical analysis with linear mixed models. Linear mixed models is a unified framework for classical random effects ANOVA models, random coefficient models and linear models for longitudinal data with associated user-friendly implementations in R and SPSS. Linear mixed models moreover provide generalizations of the classical models to complex data not covered by he standard statistical toolbox. The course will focus on modeling with mixed models, on how a statistical analysis can be carried out for a mixed model, and on interpretation of models and results. Hands-on experience with real data will be obtained through computer exercises.
Prerequisites:
A basic knowledge of statistics (linear regression) and probability theory (random variables, expectation variance and covariance) is expected.

Organizer: Professor Rasmus Waagepetersen, e-mail: rw@math.aau.dk

Lecturers: Professor Rasmus Waagepetersen

ECTS: 1

Time: 14 and 21 September 2015, from 8.15-12.00

Place: Aalborg University, Fredrik Bajers Vej 7G/5 room 109

Zip code: 9220

City: Aalborg

Number of seats: 20

Deadline: 24 August