This course is for students, researchers, and data scientists who want to move beyond predefined assessment packages and learn the principles and practice of constructing bespoke statistical assessments from the ground up. Using the latest iteration of model building tools (RTMB), which provides automatic differentiation and Laplace approximation for maximum likelihood estimation, participants will learn to translate custom scientific hypotheses into flexible, computationally efficient models. The course has a practical focus on implementation, but it will also cover statistical foundations where needed. Key topics include formulating models for non-standard data types (e.g., compositional, zero-inflated, and correlated observations), implementing random effects for hierarchical structures, and employing the Laplace approximation for fast, accurate inference. Through a series of hands-on case studies drawn from ICES assessments, attendees will gain proficiency in building, diagnosing, and critiquing their own models. By the course's conclusion, participants will be equipped with the skills and code examples to tackle common and some specialized assessment challenges with confidence, crafting tailored solutions that faithfully represent the underlying mechanisms of their systems of study.
Main target audience
Participants in ICES assessment groups. It is reccommended that participants have a basic understanding of statistics and knowledge of R will be an advantage.
If your institute requires an invoice / factura for your Training Course Registration, please click here.
Please note that issuing an invoice /factura will add a €50 administration fee to the Training Course Registration price.
