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.
Statistics
2026-09-02
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.
how regression models are estimated and judged, from MATH5806: score and fisher information, newton-raphson and the method of scoring, the wald/score/likelihood-ratio trio, and exact t and F inference in the linear gaussian model.
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.
| # | Course Code | Title | Offered | Prerequisites | Term | Type | Textbook | Notes |
|---|---|---|---|---|---|---|---|---|
| 1. | COMP6713 | Natural Language Processing | T1 | MATH1081,9444 | 26T1 | Elective | na | |
| 2. | FINS5513 | Investments and Portfolio Selection | T1,2,3 | 8750 program | 26T1 | Elective | na | |
| 3. | FINS5536 | Fixed Income Securities and Interest Rate Derivatives | T2 | 5513 | 26T2 | Elective | na | pricing, hedging, risk management. options, futures and swaps (int rate derivs) |
| 4. | MATH5856 | Introduction to Statistics and Statistical Computations | T2 | 26T2 | Elective | na | recommended for 5905 | |
| 6. | MATH5960 | Bayesian Inference and Computation | T3 | 2801/2901 | 26T3 | Elective | ||
| 7. | MATH5825 | Measure, Integration and Probability | T3 | U5705 | 26T3 | Elective | na | implicit prereq for 5835 |
| 8. | MATH5905 | Statistical Inference | T1,2,3 | U5846,U5856 | 27T1 | Core | na | |
| 9. | COMP9518 | Advanced Machine Learning | T2 | 9517 | 27T2 | Elective | na | |
| 10. | MATH5845 | Time Series | T2 | 27T2 | Elective | na | ||
| 11. | MATH5855 | Multivariate Analysis | T3 | 27T3 | Elective | na | ||
| 12. | MATH5835 | Advanced Stochastic Processes | T1 | U5825 | 28T1 | Core | na | Difficult. Requires an understanding of Real Analysis and Measure Theory |
| 13. | MATH5806 | Applied Regression Analysis | T2 | 28T2 | Elective | na | splines, poisson / binomial regression | |
| 14. | MATH5925 | Project (12uoc) | T1,2,3 | 36UoC | 28T2 | Core | na |
course pages
Per-course write-ups — admin, textbooks, and links into my notes and solutions — for the courses I have completed:
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.
Non-descriptive frisbee stats
A computer vision model that takes in streamed games and outputs a player statistic that factors in non-descriptive events — i.e. giving the correct call at the correct time, or poaching in the lane to force a bad throw.
I expect this to be trained using a transformer and written in Python. It is inspired by Andrew Wood’s analytical Ultimate dream.
Lab 2
The Chapter 2 lab is a tour of the language itself – vectors, matrices, indexing, plotting, and loading a data set. What follows is that tour, split into the sections the book uses.
Basic Commands
Vectors are built with c() (concatenate), and both <- and = assign. Arithmetic is element-wise, so operands must share a length – reassigning x to length 3 lets it line up with y:
Chapter 1: Getting Started
mean(abs(rnorm(100))) # generate 100 N(0,1) variates, take abs, then mean
rnorm(10)