Contents
The course offers a straightforward and practical approach to applied statistics using Bayesian inference for ecologists. It starts with a general introduction to statistical modeling and the concepts of Bayesian statistics (likelihood, priors, posterior distribution, MCMC sampling). We will move step-by-step from classical ANOVA and linear regression to generalized, nonlinear, or mixed-effects models, with a strong conceptual focus on the building blocks of statistical models.
While previous software required users to code in specific modeling languages (JAGS, Stan, NIMBLE), we are focusing on the user-friendly and flexible R-package ‘brms’, which makes the transition easy for people familiar with ‘lm’ or ‘lme4’. An additional introduction to coding in Stan will be provided for interested participants.
Didactic aim / competences gained
Participants learn how to practically conceptualize their research questions into statistical models. They learn how to specify and critically interpret models of varying complexity in R. The course prepares participants to analyze their own data with state-of-the-art methodology.
Prior knowledge needed
Basic knowledge in R is required (e.g. importing and transforming data). Some basic knowledge in statistics (e.g. from an introductory course) is recommended.
Participants are expected to bring their own laptops with required software installed before the course. Infos on software will be provided.
Lecturer
Dr Benjamin Rosenbaum
Benjamin is a Computational Ecologist with a PhD in Applied Mathematics, based in iDiv’s ‘Theory in Biodiversity group’. Here, his research focuses on statistical methods for species interactions and population dynamics. Through his work as a statistical advisor, he has been involved in many biodiversity-related projects across all iDiv. He also offers statistical consulting for the whole iDiv community.