Courses / Module

GENERALISED LINEAR MODELS

Module code: ST304
Credits: 5
Semester: 2
Department: MATHEMATICS AND STATISTICS
International:

	Overview
	Transformation of response and predictors, Box-Cox method, variance stabilizing transformation. Collinearity and variance inflation factors. Model selection: forwards, backwards and stepwise. All possible regressions, Mallow's Cp and PRESS. Categorical data methods. Generalized linear models, exponential families. Logistic regression for binary responses and Binomial counts. Loglinear models for Poisson counts, two- and three-way tables.

Overview

Transformation of response and predictors, Box-Cox method, variance stabilizing transformation. Collinearity and variance inflation factors. Model selection: forwards, backwards and stepwise. All possible regressions, Mallow's Cp and PRESS.

Categorical data methods. Generalized linear models, exponential families. Logistic regression for binary responses and Binomial counts. Loglinear models for Poisson counts, two- and three-way tables.

	Learning Outcomes
	On successful completion of the module, students should be able to: Build regression models. Assess need for transformations. Use model selection tools. Examine data using two- and three-way contingency tables. Fit and interpret generalised linear models.

Teaching & Learning methods

Assessment

	Autumn Supplementals/Resits
	Are supplemental/resit registrations permitted? Yes University scheduled written examination (Autumn): 120 minutes, On-campus

	Pre-Requisites
	ST303

	Timetable
	Click on the link to open the Timetable for Module [ST304]. Venue locations can be found here.

Search form

Courses / Module

GENERALISED LINEAR MODELS