Multivariate Methods

10 credits

Syllabus, Bachelor's level, 1MS003

A revised version of the syllabus is available.

Code: 1MS003
Education cycle: First cycle
Main field(s) of study and in-depth level: Mathematics G2F
Grading system: Pass with distinction (5), Pass with credit (4), Pass (3), Fail (U)
Finalised by: The Faculty Board of Science and Technology, 19 April 2010
Responsible department: Department of Mathematics

Entry requirements

60 credits including Analysis of Regression and Variance

Learning outcomes

In order to pass the course (grade 3) the student should be able to

give an account of some methods of visualizing multivariate data sets;
give an account of and use multivariate normal distribution;
perform statistical tests of the mean value vector of a multivariate normal distribution;
perform statistical tests of two or several populations of a multivariate normal distribution;
give an account of methods and techniques for validation of multivariate normal distribution;
use principal component and factor analysis for typical problems;
use canonical correlation analysis;
use classification techniques;
give an account of and use methods for multivariate cluster analysis;
present mathematical statistical arguments to others.

Content

Methods of visualisation. Multivariate normal distribution, test of mean value vector, test of one or several populations. Techniques for validation. Principal component analysis. Factor analysis. Canonical correlation analysis. Classification. Multivariate cluster analysis.

Instruction

Lectures, problem solving sessions and computer-assisted laboratory work.

Assessment

Written examination (7 credit points) at the end of the course as well as assignments, laboratory work and, possibly, oral presentation during the course (3 credit points) in accordance with instructions at course start.