GLMs, GAMs and GLMMs: an overview of theory for applications in fisheries research

This paper provides an overview of the modelling process using generalized linear models (GLMs), generalized additive models (GAMs) and generalized linear mixed models (GLMMs), especially as they are applied within fisheries research. We describe the essential aspect of model interpretation and construction so as to achieve its correct application. We start with the simplest models and show the progression from GLMs to either GAMs or GLMMs. Although this is not a comprehensive review, we emphasise topics relevant to fisheries science such as transformation options, link functions, adding model flexibility through splines, and using random and fixed effects. We finish by discussing the various aspects of these models and their variants, and provide a view on their relative benefits to fisheries research.

[1]  K. Lorenzen,et al.  Influence of Drake Passage oceanography on the parasitic infection of individual year-classes of southern blue whiting Micromesistius australis , 2003 .

[2]  P. Diggle,et al.  Analysis of Longitudinal Data , 2003 .

[3]  Han-Lin Lai,et al.  Linear mixed-effects models for weight–length relationships , 2004 .

[4]  Rory A. Fisher,et al.  256: The Analysis of Variance with Various Binomial Transformations. , 1954 .

[5]  D. A. Williams,et al.  Extra‐Binomial Variation in Logistic Linear Models , 1982 .

[6]  J. Robin,et al.  Spatio-temporal analysis of commercial trawler data using General Additive models: patterns of Loliginid squid abundance in the north-east Atlantic , 2002 .

[7]  D. A. Williams,et al.  Generalized Linear Model Diagnostics Using the Deviance and Single Case Deletions , 1987 .

[8]  T. Quinn.  Catch-Per-Unit-Effort: A Statistical Model for Pacific Halibut (Hippoglossus stenolepis) , 1985 .

[9]  Carl de Boor,et al.  A Practical Guide to Splines , 1978, Applied Mathematical Sciences.

[10]  I. D. Tuck,et al.  The influence of sediment type on the distribution of the lesser sandeel, Ammodytes marinus , 2000 .

[11]  Statistical Models for Estimating CPUE from Catch and Effort Data , 1992 .

[12]  D. J. Finney.  On the Distribution of a Variate Whose Logarithm is Normally Distributed , 1941 .

[13]  M. Ortiz,et al.  Alternative error distribution models for standardization of catch rates of non-target species from a pelagic longline fishery: billfish species in the Venezuelan tuna longline fishery , 2004 .

[14]  D. Cox,et al.  A General Definition of Residuals , 1968 .

[15]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[16]  Gunnar Stefánsson,et al.  Analysis of groundfish survey abundance data: combining the GLM and delta approaches , 1996 .

[17]  David Firth,et al.  Multiplicative Errors: Log‐Normal or Gamma? , 1988 .

[18]  Grace Wahba,et al.  Spline Models for Observational Data , 1990 .

[19]  A. Punt,et al.  Standardization of catch and effort data in a spatially-structured shark fishery , 2000 .

[20]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[21]  John M. Hoenig,et al.  Direct estimates of gear selectivity from multiple tagging experiments , 1997 .

[22]  D. Butterworth,et al.  Using a GLMM to estimate the somatic growth rate trend for male South African west coast rock lobster , 2004 .

[23]  P. McCullagh,et al.  Generalized Linear Models, 2nd Edn. , 1990 .

[24]  R. Schall.  Estimation in generalized linear models with random effects , 1991 .

[25]  John A. Nelder,et al.  Generalized linear models. 2nd ed. , 1993 .

[26]  C. C. Heyde,et al.  Quasi-Likelihood and Optimal Estimation, Correspondent Paper , 1987 .

[27]  R. Wolfinger,et al.  Generalized linear mixed models a pseudo-likelihood approach , 1993 .

[28]  D. Squires,et al.  Skipper skill and panel data in fishing industries , 1999 .

[29]  N. Breslow,et al.  Approximate inference in generalized linear mixed models , 1993 .

[30]  Nan M. Laird,et al.  Longitudinal panel data: an overview of current methodology , 1996 .

[31]  P. J. Bromley,et al.  Growth, sexual maturation and spawning in central North Sea plaice (Pleuronectes platessa L.), and the generation of maturity ogives from commercial catch data , 2000 .

[32]  David Firth,et al.  On the efficiency of quasi-likelihood estimation , 1987 .

[33]  W. Venables,et al.  A generalised linear model for catch allocation: an example from Australia's Northern Prawn Fishery , 2004 .

[34]  André E. Punt,et al.  Stock assessment of the blue grenadier Macruronus novaezelandiae resource off south-eastern Australia , 2001 .

[35]  I. Priede,et al.  Improving the precision of the daily egg production method using generalized additive models , 1997 .

[36]  K. Bigelow,et al.  Environmental effects on swordfish and blue shark catch rates in the US North Pacific longline fishery , 1999 .

[37]  M. Cardinale,et al.  The influence of stock structure and environmental conditions on the recruitment process of Baltic cod estimated using a generalized additive model , 2000 .

[38]  Pierre Kleiber,et al.  Generalized additive model and regression tree analyses of blue shark (Prionace glauca) catch rates by the Hawaii-based commercial longline fishery , 2001 .

[39]  Jenný Brynjarsdóttir,et al.  Analysis of cod catch data from Icelandic groundfish surveys using generalized linear models , 2004 .

[40]  J. Lewis,et al.  Probit Analysis (3rd ed). , 1972 .

[41]  A. Punt,et al.  Standardizing catch and effort data: a review of recent approaches , 2004 .

[42]  C. Legault,et al.  All alternative method for estimating bycatch from the U.S. shrimp trawl fishery in the Gulf of Mexico, 1972-1995 , 2000 .

[43]  S. Bannerot,et al.  Using Frequency Distributions of Catch per Unit Effort to Measure Fish‐Stock Abundance , 1983 .

[44]  D. Kimura.  Standardized measures of relative abundance based on modelling log (c.p.u.e.), and their application to Pacific ocean perch (Sebastes alutus) , 1981 .

[45]  R. W. Wedderburn.  Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method , 1974 .

[46]  B. Wiens.  When Log-Normal and Gamma Models Give Different Results: A Case Study , 1999 .

[47]  David G. Kendall,et al.  Spline Transformations: Three New Diagnostic Aids for the Statistical Data‐Analyst , 1971 .

[48]  G. Robinson.  That BLUP is a Good Thing: The Estimation of Random Effects , 1991 .

[49]  R. Tibshirani,et al.  Generalized Additive Models , 1991 .