2026/27 Taught Postgraduate Module Catalogue

MATH3702 Multivariate Analysis and Classification

20 Credits Class Size: 200

Module manager: TBC
Email: TBC

Taught: Semester 2 (Jan to Jun) View Timetable

Year running 2026/27

Pre-requisite qualifications

None

Pre-requisites

MATH2701 Statistical Methods

Mutually Exclusive

MATH3772 Multivariate Analysis
MATH5772M Multivariate&Cluster Analysis

Module replaces

MATH3772 Multivariate Analysis MATH5772 Multivariate and Cluster Analysis

This module is not approved as an Elective

Module summary

In today’s data-driven world, almost every data set is multivariate. It is typical that experimental units or individuals are measured on more than one variable at a time. The purpose of multivariate analysis is to develop methods to uncover underlying structures and patterns in complex multivariate data to enhance the understanding and interpretation of how multiple variables interact in multivariate data.

Objectives

This module will teach students the theory and methods to analyse and interpret multivariate data. Some methods, such as estimation and testing for the multivariate normal distribution, can be viewed as extensions of methods for univariate continuous data. Other methods are specific to the multivariate setting. These include methods for dimension reduction (principal component analysis, factor analysis, and multidimensional scaling) and methods to classify individuals into groups (discriminant analysis, cluster analysis, and classification and regression trees). Computer software will be used to perform analyses and communicate results with applications to real data.

Learning outcomes

On successful completion of the module students will be able to: 1. summarize multivariate data in terms of the mean vector and covariance/correlation matrices 2. construct joint, marginal and conditional distributions, with particular emphasis on the multivariate normal distribution; 3. obtain and use Hotelling's T2 statistic for one-sample and two-sample testing problems; 4. motivate and carry out principal component analysis and factor analysis to reduce the number of variables in multivariate data; 5. derive and interpret decision rules in discriminant analysis; 6. partition data into clusters using mixture models and hierarchical methods; 7. use multidimensional scaling to construct low-dimensional representations of data; 8. use decision trees and random forests to uncover underlying patterns in classification and regression 9. analyse real data using a statistical package and write a report giving and interpreting the results.

Syllabus

The details of the syllabus and learning activities will be published in the Minerva organisation, or other student facing information for the module, and available to students once they are enrolled.

Teaching Methods

Delivery type Number Length hours Student hours
Lectures 44 1 44
Practicals 1 2 2
Private study hours 154
Total Contact hours 46
Total hours (100hr per 10 credits) 200

Opportunities for Formative Feedback

Formative feedback will be provided on regular example sets or other similar learning activity.

Reading List

Check the module area in Minerva for your reading list

Last updated: 30/04/2026

Errors, omissions, failed links etc should be notified to the Catalogue Team