Workshop on Generalized Linear Mixed Models: Concepts and Applications

19-22.3.2018 in Lapland Hotels Luostotunturi, Luostontie 1, Luosto, Finland


Welcome to the statistical highlight of the decade in Finland! We are proud to announce that Professor Walter W. Stroup is arriving to Finland to give a workshop on Generalized Linear Mixed Models in Luosto Resort, Northern Finland. Generalized linear mixed models cover a very broad range of models that are widely used in many disciplines. “Generalized” refers to the models being able to handle data arisen from several different distributions, not just the normal distribution. And the “Mixed” refers to the models being able to handle correlated observations due, e.g., to temporal or spatial structures through so called random effects. This workshop gives us practical information and tools to apply these models to actual data. The workshop is organized by Natural Resources Institute Finland (Luonnonvarakeskus, Luke), but it is open for everybody to attend (see information about registration fees below). We are planning to organize a full scientific program combined with opportunities to enjoy the amazing nature in Luosto and the emerging spring in the Finnish Lapland. The workshop is supported by the Academic Program of SAS Finland.


DESCRIPTION: This three-day course is intended for those who want to learn about the generalized linear mixed models (GLMM) for data analysis and GLMM-based tools for planning research studies. The course will consist of two days of lecture and one day of hands-on interactive demonstrations, with each day consisting of a morning session and an afternoon session. The material is presented at an applied level, accessible to participants with training in linear statistical models.

On the first day, the morning session will cover fundamental GLMM concepts essential to understanding the applications that will follow in subsequent sessions. The afternoon session will cover common distributions and their associated modeling, estimation and inference strategies. On the second day, the morning session will cover GLMM methodology for predictions and GLMMs for repeated measures in time and space (longitudinal and spatial data). The afternoon session will begin with planning, power and sample size methodology for models covered in the previous sessions. The afternoon session will continue with a review of what we know (and don’t know) about best practices when using GLMMs, and how we use simulation to further our understanding of best practices. The afternoon session concludes with GLMM-based variable selection strategies. On the third day, we will have an interactive lab with demonstrations that allow workshop participants to work with various examples, and try things of interest but not covered in the lecture sessions. The third day will also feature a talk about statistics in journal publications from a non-statistician “consumer of statistical methods” perspective. 

Computations for all workshop examples use mixed model tools in SAS®.


From the MENU (top of a page) you can find schedule, practical info and registration forms!

Registration deadline 30.11.!