Dear all,
We are excited to announce our upcoming course on (GENERALISED) LINEAR MIXED MODELS IN R, taking place from October 9th to 13th, 2023. To ensure global accessibility, this course will be conducted online, allowing participants from all over the world to benefit from the invaluable knowledge shared.
COURSE OVERVIEW:
The (GENERALISED) LINEAR MIXED MODELS (GLMMs) framework has become the gold standard for statistical analysis of experimental and observational data. This course focuses on extending standard LM/GLM models with random effects, variance structures, and correlation structures to account for grouped data, non-constant variances (heteroscedasticity), and residual spatial, temporal, and phylogenetic autocorrelation. Participants will gain practical skills in specifying, interpreting, and validating linear and generalized linear mixed models, with a specific emphasis on the lme4 and glmmTMB regression packages in R.
TARGET AUDIENCE AND ASSUMED BACKGROUND:
This course is designed for graduate students and researchers who already possess experience with generalized linear regression models in R and wish to enhance their understanding by incorporating random effects, correlation structures, and variance models. While a review of LM, GLM, and ANOVA basics is provided at the beginning, those without a solid foundation in these topics may find our course "Generalized Linear Models as a unified framework for data analysis in R" more suitable.
LEARNING OUTCOMES:
- Deepen understanding of fundamental regression concepts, including centering, scaling, interactions, contrasts, and ANOVA.
- Gain insight into the components of the GLMM framework, such as distribution choices, random effects, and correlation structures.
- Develop the ability to select the appropriate model structure for applied analysis of experimental or observational data, with a focus on lme4 and glmmTMB packages.
- Acquire visualization techniques for fitted GLMMs using the effects package and learn to assess model assumptions using the DHARMa package.
PROGRAM:
Sessions will run from 14:00 to 20:00 (Berlin time) Monday through Thursday, with a shorter session from 14:00 to 18:00 on Friday. Our comprehensive program incorporates a mix of lectures, in-class discussions, Q&A sessions, and practical exercises conducted through Slack and Zoom.
Please find below a brief outline of the course schedule:
- Monday: Regression basics, linear model interpretation, multiple regression, scaling, centering, omitted-variable bias, interactions, contrasts for categorical predictors, and ANOVA.
- Tuesday: Introduction to random effects, variance structures (GLS), model selection, and causal inference.
Wednesday: Merging variance structures, random effects, and the GLM framework to work with GLMMs. Specification, diagnostics, common issues, variance partitioning, and model selection.
Thursday: Exploring correlation structures beyond random effects, including spatial, temporal, and phylogenetic structures, and their incorporation into regression models.
- Friday: Advanced topics, including non-parametric methods, null models, bootstrap, cross-validation, Bayesian estimation vs. frequentist estimation of GLMMs, and structural equation models (emphasis based on group interest).
For more details and registration, please visit our course page: https://www.physalia-courses.org/courses-workshops/glmms-in-r/
www.physalia-courses.org
