Generalized linear models

Buckley, Yvonne M. (2015). Generalized linear models. In Gordon A. Fox, Simoneta Negrete-Yankelevich and Vinicio J. Sosa (Ed.), Ecological statistics: contemporary theory and application (pp. 131-148) Oxford, United Kingdom: Oxford University Press. doi:10.1093/acprof:oso/9780199672547.003.0007

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Buckley, Yvonne M.
Title of chapter Generalized linear models
Title of book Ecological statistics: contemporary theory and application
Place of Publication Oxford, United Kingdom
Publisher Oxford University Press
Publication Year 2015
Sub-type Research book chapter (original research)
DOI 10.1093/acprof:oso/9780199672547.003.0007
Open Access Status Not Open Access
ISBN 9780199672554
9780199672547
Editor Gordon A. Fox
Simoneta Negrete-Yankelevich
Vinicio J. Sosa
Chapter number 6
Start page 131
End page 148
Total pages 18
Total chapters 13
Collection year 2016
Language eng
Abstract/Summary Generalized Linear Modeling (GLM) unifies several statistical techniques, providing a stable and modular foundation on which to build a useful working knowledge of statistical modeling. GLMs enable a re-interpretation of previously learned ANOVA and regression techniques and integrate well with more advanced modeling techniques introduced in later chapters. This chapter describes the modular components of a GLM, the linear predictor, error distribution, and link function, and through explanation and examples guides readers to develop a more intuitive understanding of the nature of their data set and the generalized linear models that can be used to make sense of it. Ecologists commonly collect data that contravene the assumptions of ANOVA and regression; GLMs provide a more flexible modeling framework that enables analysis of binary, count, and proportion data, as well as continuous data with non-normal error structures. These data can also be challenging to plot and interpret; examples are provided throughout with associated code for graphing data and the models developed. The chapter also provides guidance on model criticism and inference. Finally there are a number of reasons why your GLM may not work or may be inappropriate, and this chapter discusses how to recognize problems and suggest solutions.
Keyword Binomial
Poisson
Binary
Non-normal
Maximum likelihood
Error distribution
Count
Proportion
Q-Index Code B1
Q-Index Status Confirmed Code
Institutional Status UQ

Document type: Book Chapter
Collections: Official 2016 Collection
School of Biological Sciences Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 1 times in Thomson Reuters Web of Science Article | Citations
Google Scholar Search Google Scholar
Created: Sun, 28 Jun 2015, 00:46:45 EST by System User on behalf of Scholarly Communication and Digitisation Service