A student's guide to MATH5806 at UNSW: generalised linear models in R, the Dobson & Barnett textbook, assessment breakdown (quiz, R assignment, mid-session, final), plus free notes and fully worked GLM solutions.
Glm
2026-09-08
Annette J. Dobson and Adrian G. Barnett, An Introduction to Generalized
Linear Models, 4th edition, Chapman & Hall/CRC, 2018. The MATH5806 text:
exponential family, estimation and inference, normal linear models,
binomial and Poisson regression, contingency tables, survival analysis,
clustered and longitudinal data, Bayesian methods and MCMC. The book’s
datasets ship in the dobson R package.
Solutions to every exercise live at Solutions to Dobson & Barnett’s An Introduction to Generalized Linear Models.
exponential-family GLM theory from MATH5806: the canonical form, link functions, fisher scoring as iteratively reweighted least squares, deviance, poisson and log-linear regression, and quasi-likelihood.
logistic regression developed honestly as a GLM, from MATH5806: bernoulli in canonical form, the logit link, IRLS from scratch in R, deviance and odds-ratio inference, separation, and the machine-learning reading as corollary.
the regression shelf: ordinary least squares, regularised and locally-weighted variants, logistic regression, and the generalised linear model theory that unifies them — estimation, inference, and worked implementations.
Worked solutions in R to all 78 exercises in the fourth edition of Dobson & Barnett's An Introduction to Generalized Linear Models — exponential family theory, estimation and inference, normal linear models, binomial and Poisson regression, contingency tables, survival analysis, clustered and longitudinal data, and Bayesian analysis with MCMC.