- ホーム
- > 洋書
- > 英文書
- > Science / Mathematics
Full Description
Bayesian Analysis of Capture-Recapture Data with Hidden Markov Models: Theory and Case Studies in R and NIMBLE introduces ecologists and statisticians to a powerful and unifying framework for analysing capture-recapture data. Hidden Markov models (HMMs) have become a cornerstone in modern population ecology, offering a flexible way to decompose complex processes such as survival, recruitment, and dispersal into simpler building blocks, while explicitly accounting for the fact that we only observe imperfect data rather than the true underlying states. Combined with Bayesian inference, HMMs provide a natural and transparent approach to handle uncertainty, explore model structures, and draw robust conclusions. This book illustrates how to bring these ideas to life using the R package NIMBLE, a fast-developing environment for building and fitting hierarchical models.
Key features include:
• A clear introduction to the principles of Bayesian statistics, HMMs, and the NIMBLE package
• Step-by-step tutorials showing how to implement a wide range of capture-recapture models for open populations
• Fully reproducible examples with data and R code, following a "learning by doing" philosophy
• Case studies drawn from the ecological literature, illustrating how to apply methods to real-world conservation questions
• Practical guidance on model specification, coding strategies, and interpretation of results
Written in an accessible style, this book is designed for ecologists, wildlife biologists, and conservation scientists who already use R and wish to deepen their modelling toolkit, as well as statisticians interested in ecological applications. Beginners will find a self-contained path into Bayesian capture-recapture modelling, while experienced researchers will discover a flexible framework to extend and adapt to their own data and questions.
Contents
1. Bayesian statistics & MCMC. 2. NIMBLE tutorial. 3. Hidden Markov models. 4. Alive and dead. 5. Sites and states. 6. Dealing with covariates. 7. Addressing model lack of fit. 8. Quantifying life history traits.
