School of Mathematics and StatisticsHome | About the school | Contact | Courses | Research | Personnel list
MT4609 MULTIVARIATE ANALYSIS
AimsTo introduce students to the ideas and techniques of multivariate statistical analysis.
ObjectivesBy the end of the course students are expected to be able to
- reproduce the underlying theory;
- estimate parameters and carry out appropriate tests;
SyllabusMean vectors, covariance matrices, correlation matrices; basic properties of multivariate normal distributions; checking multivariate normality; the likelihood ratio and union-intersection principles for constructing multivariate tests; the one-sample and two-sample Hotelling's T2 tests; tests on covariance matrices, tests of independence; linear discriminant analysis; principal components analysis; canonical correlation.
TextbooksMultivariate Statistical Methods: D F Morrison; Thomson/Brooks/Cole; (4th edition) 2005.
Applied Multivariate Analysis: R A Johnson & D W Wichern; Prentice-Hall; (6th edition) 2007;
Multivariate Analysis: K V Mardia, J T Kent & J M Bibby; Academic Press; 1979;
Assessment2 Hour Examination = 100%
AvailabilityAcademic year 2012/13 in semester 2 at 10
LecturerDr I B J Goudie
Click here for access to past examination papers via iSaint.
Click here to see the University Course Catalogue entry.
Revised: PMH (October 2012)