Applied Statistical Modelling for Ecologists
A Practical Guide to Bayesian and Likelihood Inference Using R, JAGS, NIMBLE, Stan and TMB
- 520 pages
- English
- PDF
- Only available on web
Applied Statistical Modelling for Ecologists
A Practical Guide to Bayesian and Likelihood Inference Using R, JAGS, NIMBLE, Stan and TMB
About This Book
Applied Statistical Modelling for Ecologists provides a gentle introduction to the essential models of applied statistics: linear models, generalized linear models, mixed and hierarchical models. All models are fit with both a likelihood and a Bayesian approach, using several powerful software packages widely used in research publications: JAGS, NIMBLE, Stan, and TMB. In addition, the foundational method of maximum likelihood is explained in a manner that ecologists can really understand. This book is the successor of the widely used Introduction to WinBUGS for Ecologists (Kéry, Academic Press, 2010). Like its parent, it is extremely effective for both classroom use and self-study, allowing students and researchers alike to quickly learn, understand, and carry out a very wide range of statistical modelling tasks. The examples in Applied Statistical Modelling for Ecologists come from ecology and the environmental sciences, but the underlying statistical models are very widely used by scientists across many disciplines. This book will be useful for anybody who needs to learn and quickly become proficient in statistical modelling, with either a likelihood or a Bayesian focus, and in the model-fitting engines covered, including the three latest packages NIMBLE, Stan, and TMB.
- Contains a concise and gentle introduction to probability and applied statistics as needed in ecology and the environmental sciences
- Covers the foundations of modern applied statistical modelling
- Gives a comprehensive, applied introduction to what currently are the most widely used and most exciting, cutting-edge model fitting software packages: JAGS, NIMBLE, Stan, and TMB
- Provides a highly accessible applied introduction to the two dominant methods of fitting parametric statistical models: maximum likelihood and Bayesian posterior inference
- Details the principles of model building, model checking and model selection
- Adopts a "Rosetta Stone" approach, wherein understanding of one software, and of its associated language, will be greatly enhanced by seeing the analogous code in other engines
- Provides all code available for download for students, at https://www.elsevier.com/books-and-journals/book-companion/9780443137150
Frequently asked questions
Information
Table of contents
- Front Cover
- Applied Statistical Modelling for Ecologists
- Copyright Page
- Dedication
- Contents
- Foreword
- Acknowledgments
- 1 Introduction
- 2 Introduction to statistical inference
- 3 Linear regression models and their extensions to generalized linear, hierarchical, and integrated models
- 4 Introduction to general-purpose model fitting engines and the “model of the mean”
- 5 Normal linear regression
- 6 Comparing two groups in a normal model
- 7 Models with a single categorical covariate with more than two levels
- 8 Comparisons along two classifications in a model with two factors
- 9 General linear model for a normal response with continuous and categorical explanatory variables
- 10 Linear mixed-effects model
- 11 Introduction to the generalized linear model (GLM): comparing two groups in a Poisson regression
- 12 Overdispersion, zero inflation, and offsets in a Poisson GLM
- 13 Poisson GLM with continuous and categorical explanatory variables
- 14 Poisson generalized linear mixed model, or Poisson GLMM
- 15 Comparing two groups in a logistic regression model
- 16 Binomial GLM with continuous and categorical explanatory variables
- 17 Binomial generalized linear mixed model
- 18 Model building, model checking, and model selection
- 19 Occupancy models
- 20 Integrated models
- 21 Conclusion
- Bibliography
- Index
- Back Cover