You’re using AIC wrong! Causal inference vs model selection in biodiversity science

Workshop

You’re using AIC wrong! Causal inference vs model selection in biodiversity science

Tuesday, December 10 2024 01:00PM-05:00PM

Hybrid at McGill

Language of the workshop: EN & FR

Ecologists have two broad sets of tools in front of them. One set of tools, including AIC, aims to help us make good predictions and forecasts about other observations. Another set of tools helps us understand our systems and explain why things change, and how our interventions might cause that change. These second set of tools are called causal inference. Problems often result from using tools meant for prediction when what we really want is causation.

Causal inference starts with defining what variables influence which others. From there we can learn if we have the data we need to answer our questions, which models are useful, and how to measure our effect when we have it. We’ll learn how to draw directed acyclic graphs, when to control for variables (and when NOT to control for them!), and do some practical examples.

REGISTRATION