ST2LM-Linear Models

Module Provider: Mathematics and Statistics
Number of credits: 10 [5 ECTS credits]
Level:5
Terms in which taught: Autumn term module
Pre-requisites: ST1PS Probability and Statistics
Non-modular pre-requisites:
Co-requisites:
Modules excluded:
Current from: 2021/2

Module Convenor: Dr Karen Poulter
Email: k.l.poulter@reading.ac.uk

Type of module:

Summary module description:
This module covers the most common models used in statistics: multiple linear regression for observational studies and completely randomised designs for planned studies.

Aims:
Linear models are used widely in statistics. The most common models will be reviewed and their relationship to the general linear model explored. The main aim of the module is to present a standard approach for fitting linear models to data and for comparing alternative linear models with one another. The module also aims to provide the skills to develop and test linear models appropriate for a range of practical problems.

Assessable learning outcomes:

On completion of this module students will have acquired:

• knowledge of the theory associated with the general linear model and the principles of analysis of variance;


• the ability to fit regression models to data, interpret them and check their adequacy;


• an awareness of the role of regression modelling in the analysis of data from designed experiments;


• the ability to use SAS to fit linear models and check their ad equacy.

Additional outcomes:

Outline content:

Simple linear regression and the completely randomised design.

The General Linear Model for multiple regression: definition and matrix notation. Model checking: residual analysis, influential observations.

Further topics: variable selection, models with continuous and categorical variables, polynomial regression and multicollinearity.

Use of SAS and R to fit linear models.

Brief description of teaching and learning methods:
Lectures, supported by problem sheets, PC practicals and tutorials.