Zero-Inflated Models A one-day live online course on zero-inflated models in R. Learn to model count data with excess zeros using ZIP, ZINB, and hurdle approaches, plus model diagnostics and interpretation.
https://www.prstats.org/course/zero-inflated-models-zimr01/ *Master Zero-Inflated Models: a One-Day Workshop in R* Count data with lots of zeros show up everywhere — in ecology, epidemiology, public health, insurance — and simple Poisson or negative-binomial models often fail. *Zero-Inflated Models (ZIMR01)* is a focused, one-day live online course that gives you both theory and hands-on practice to model such data correctly. ------------------------------ Why this course is essential - Real datasets often exhibit *excess zeros or overdispersion*, which can mislead inference if ignored. - You’ll learn to diagnose when zero inflation is present (versus mere dispersion), using simulation-based residual tools. - You’ll see how to fit, compare, and interpret *Zero-Inflated Poisson (ZIP)* and *Zero-Inflated Negative Binomial (ZINB)* models, and understand when *hurdle* or *truncated* models are more appropriate. - The skills you gain apply across fields — from ecology to health to insurance — whenever count data with many zeros arise. ------------------------------ What you will get - A grounding in Poisson, negative binomial, generalized Poisson, and Conway–Maxwell Poisson models - Tools for diagnosing overdispersion/zero inflation using the *DHARMa* package - Implementation of ZIP, ZINB, hurdle and zero-truncated models in R (e.g. via pscl::zeroinfl, glmmTMB) - Opportunities to bring your own data (time permitting) for discussion and application - Code, datasets, slides, and 30 days of post-course email support and recording access ------------------------------ Logistics & audience - *Duration*: 1 day, 6 hours - *Next date*: 1 December 2025 - *Format*: Live online (all sessions recorded) - *Fee*: £150 - *Audience*: Ecologists, epidemiologists, data analysts, early-career researchers, postgraduate students — anyone working with count data and zero inflation. - *Prerequisites*: Basic familiarity with R and elementary statistical concepts (linear regression, GLMs) is assumed. ------------------------------ If you deal with count data and want reliable methods rather than ad hoc fixes, this course equips you with the theory, implementation skills, and diagnostic tools to model zeros properly. Email [email protected] with any questions. -- Oliver Hooker PhD. PR stats [[alternative HTML version deleted]] _______________________________________________ R-sig-ecology mailing list [email protected] https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
