STAT3030Generalised Linear Models

Course handbook

Description

How do we model data of very different types in a consistent way? This course explores generalized linear models and illustrates how methods for analysing continuous and categorical data fit into this framework.

Availability2024 Course Timetables

Learning outcomes

On successful completion of the course students will be able to:

1. Understand and use the principles of statistical modelling;

2. Have a unified conceptual and theoretical framework for many of the most commonly used statistical methods including multiple linear regression, analysis of variance and logistic regression;

3. Develop skills in statistical computing, specifically in the R statistical programming language and R graphics.

4. Write up a report/project on an analysis of a data set and provide a clear report of results with critical interpretation as based on R code, R outputs and theoretical understanding of theory given in the lectures.

Content

Topics include:

• Linear models for continuous data (regression and ANOVA)
• Model fitting as an approach to statistical analysis
• Least squares estimation
• Maximum likelihood estimation
• Inference methods based on model fitting
• Exponential family of distributions
• Generalised Linear Models
• Models for categorical data (logistic regression for nominal and ordinal data, Poisson regression and log-linear models)
• Generalised Additive models

Assumed knowledge

STAT1300 Fundamentals of Statistics, STAT2000 Applied Statistics and Research Methods

Assessment items

Written Assignment: Assignment 1

Written Assignment: Assignment 2

Presentation: Oral for Project

Project: Written Project

Formal Examination: Examination - Formal

Contact hours

Semester 1 - 2024 - Callaghan

Course outline

• STAT3030 - Semester 1, 2024 (Callaghan) (PDF, 187.9 KB)