Categorical Data Analysis and Generalised Linear Models

Description

This course presents methods for analyses of categorical data, with a key focus on generalised linear models (GLMs). Students will study contingency tables, and gain experience with a range of generalised linear models appropriate to binary, count, nominal, ordinal and time-to-event response variables. There will be an emphasis on understanding the theoretical basis for the models, gaining practical experience with model fitting, checking model assumptions and communicating results to non-statistical colleagues.

Availability

WebLearn GradSchool

  • Semester 2 - 2017

Learning Outcomes

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

1. Successfully analyse contingency tables using a range of standard approaches, including methods for stratified and matched data;

2. Understand the theoretical basis for generalised linear models and approaches to estimation for GLMs;

3. Appropriately select, fit and interpret results from GLMs for binary, ordinal and nominal data;

4. Appropriately analyse count data using robust Poisson regression and check whether standard distributional assumptions are met;

5. Analyse time-to-event data using Cox regression;

6. 6. Interpret results for a range of GLMs. 6. Interpret results for a range of GLMs. Interpret results for a range of GLMs.

Content

Topics covered include contingency tables, the exponential family and generalised linear models, estimation and modelling using logistic regression, log-linear models, Poisson regression, logit and probit models, multinomial models and Cox regression.  There will be an emphasis on fitting appropriate models using SAS statistical software, checking model fit and interpreting results.

 

 

 

Assessment Items

Written Assignment: Assignment 1

Written Assignment: Assignment 2

Written Assignment: Assignment 3

Written Assignment: Assignment 4

Contact Hours

WebLearn GradSchool

Online Activity

Online 6 hour(s) per Week for Full Term

Contact Hours are an Indication Only