作者: Elena N. Ieno , Anatoly A. Saveliev , Alain F. Zuur
DOI:
关键词: Markov chain Monte Carlo 、 Mixture model 、 Mathematics 、 Generalized linear mixed model 、 Statistics 、 Poisson distribution 、 Negative binomial distribution 、 Random effects model 、 Overdispersion 、 Econometrics 、 Bayesian statistics
摘要: Chapter 1 provides a basic introduction to Bayesian statistics and Markov Chain Monte Carlo (MCMC), as we will need this for most analyses. If you are familiar with these techniques suggest quickly skimming through it. In 2 analyse nested zero inflated data of sibling negotiation barn owl chicks. We explain application Poisson GLMM 1-way discuss the observation-level random intercept allow overdispersion. show that introduce GLMM. recommend reading chapter in detail, refer often Data sandeel otolith presence seal scat is analysed 3. present flowchart steps selecting appropriate technique: GLM, negative binomial or GAM, GLMs distribution. 4 relevant readers interested analysis (zero inflated) 2-way data. The takes us marmot colonies: multiple colonies animals sampled repeatedly over time. Chapters 5 - 7 address spatial correlation. presents an Common Murre density introduces hurdle models using GAM. Random effects used model 6 skate abundance recorded at approximately 250 sites along coastal continental shelf waters Argentina. also involves correlation (parrotfish abundance) collected around islands, which increases complexity analysis. residual conditional auto-regressive structures used. 8 apply click beetle 9 inflation, temporal auto-correlation. time series whale strandings. 10 demonstrate excessive number zeros does not necessarily mean inflation. whether mixture requires include false algorithm can indicate false.