Pre-requisite modules for Natural Sciences students: Statistics Option MAT00033I.
Module will run
Occurrence
Teaching period
A
Spring Term 2022-23
Module aims
To introduce the main ideas of multivariate statistical analysis; that is, the analysis of sets of data where there are several measurements on each of a number of individuals.
Module learning outcomes
A knowledge and understanding of models and methods for multivariate data.
A reasonable degree of familiarity with some of the main techniques of multivariate analysis.
Apply appropriate techniques to different sets of data.
Use the statistical package R to analyse multivariate data by various techniques.
Module content
Syllabus
Introduction: Aims of multivariate analysis, descriptive statistics, graphical representation, basic concepts of vectors and matrices, use of the R program for matrix algebra and multivariate analysis.
The Multivariate Normal Distribution: Properties of the multivariate normal, contours of constant density, marginal and conditional distribution, checking normality.
Hotelling's T-squared test:One-sample tests, two-sample tests, large sample inference.
Multvariate Analysis of Variance (MANOVA): One-way and two-way MANOVA, Wilks' Lambda and other criteria.
Principal component analysis: Principal components, principle component analysis by correlation matrix, choosing the number of components.
Factor analysis: The idea of factor analysis, estimation of loadings, choosing the number of factors.
Current Department policy on feedback is available in the undergraduate student handbook. Coursework and examinations will be marked and returned in accordance with this policy.
Indicative reading
Richard Johnson, Dean Wichern, Applied Multivariate Statistical Analysis, Prentice Hall.
H R Neave, Statistics Tables for Mathematicians, Engineers, Economists and the Behavioural and Management Sciences, Routledge.