Chapter 16 - Poisson Mixed-Effects Model (Poisson GLMM)

Abstract

The Poisson generalized linear mixed model (GLMM) is an extension of the Poisson generalized linear model (GLM) to include at least one additional source of random variation over and above the random variation intrinsic to a Poisson distribution. It adopts a Poisson GLMM to analyze a set of long-term population surveys of Red-backed shrikes. It analyzes the shrike counts under the assumption that all shrike populations have the same trend, but at different levels, corresponding to a random-intercepts model. Both frequentist and Bayesian analyses are fairly straightforward, but the model fitted in WinBUGS appears more transparent.