Overview

MATH5855 is a Post Graduate Mathematics course also available as an Honours Course. See the course overview below.

Units of credit: 6

Pre-requisites:  UNSW UG students - MATH2801, MATH2901or admitted to the postgraduate programs of the School of Mathematics and Statistics and as an elective in some approved programs.

Cycle of offering: Term 3

Graduate attributes: The course will enhance your research, inquiry and analytical thinking abilities.

More information: The course outline for 2023 will be linked within the course offering table for the term applicable. Outlines will be made available prior to the start of the applicable term. 

The online handbook entry contains information about the course. The timetable is only up-to-date if the course is being offered this year.

If you are currently enrolled in MATH5855, you can log into UNSW Moodle for this course.

Course overview

Multivariate statistical analysis is performed with the aim to encompass the data concerning all variables into one analysis. This allows for a better and deeper investigation of the relationships between the variables in comparison to the piecemeal analyses of portions of the data.

It also requires more advanced mathematical and computational techniques in comparison to the univariate analysis but the effort pays off in many ways in the quality of the resulting statistical analysis. Most multivariate methods are easier to be described and discussed under the assumption of multivariate normal distribution of the data and this will be the starting point of the course. We shall discuss likelihood ratio tests for multivariate means, for covariance matrices and covariance structures.

Estimation and testing aspects of correlations, partial correlations, and multiple correlations will be studied then. Important practical applications such as discriminant analysis, cluster analysis, principal components, canonical analysis, factor analysis and latent variables will be discussed in detail. SAS-based Computing features prominently in the course.