Show simple item record

dc.contributor.advisorKoen, Chris
dc.contributor.authorSiluyele, Ian John
dc.date.accessioned2024-02-19T10:04:41Z
dc.date.available2024-02-19T10:04:41Z
dc.date.issued2007
dc.identifier.urihttp://hdl.handle.net/11394/10652
dc.description>Magister Scientiae - MScen_US
dc.description.abstractThe multivariate two-sample tests provide a means to test the match between two multivariate distributions. Although many tests exist in the literature, relatively little is known about the relative power of these procedures. The studies reported in the thesis contrasts the effectiveness, in terms of power, of seven such tests with a Monte Carlo study. The relative power of the tests was investigated against location, scale, and correlation alternatives. Samples were drawn from bivariate exponential, normal and uniform populations. Results from the power studies show that there is no single test which is the most powerful in all situations. The use of particular test statistics is recommended for specific alternatives. A possible supplementary non-parametric graphical procedure, such as the Depth-Depth plot, can be recommended for diagnosing possible differences between the multivariate samples, if the null hypothesis is rejected. As an example of the utility of the procedures for real data, the multivariate two-sample tests were applied to photometric data of twenty galactic globular clusters. The results from the analyses support the recommendations associ­ated with specific test statistics.en_US
dc.language.isoenen_US
dc.publisherUniversity of the Western Capeen_US
dc.subjectPermutation methoden_US
dc.subjectNon-parametric testen_US
dc.subjectEuclidean distanceen_US
dc.subjectMultivariate two-sample testen_US
dc.subjectNearest neighbour testsen_US
dc.titlePower studies of multivariate two-sample tests of comparisonen_US
dc.typeThesisen_US
dc.rights.holderUniversity of the Western Capeen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record