Show simple item record

dc.contributor.authorHindberg, Kristian
dc.contributor.authorHannig, Jan
dc.contributor.authorGodtliebsen, Fred
dc.date.accessioned2019-10-14T11:24:12Z
dc.date.available2019-10-14T11:24:12Z
dc.date.issued2019-01-22
dc.description.abstractTwo classical multivariate statistical problems, testing of multivariate normality and the <i>k</i>-sample problem, are explored by a novel analysis on several resolutions simultaneously. The presented methods do not invert any estimated covariance matrix. Thereby, the methods work in the High Dimension Low Sample Size situation, i.e. when <i>n</i> ≤ <i>p</i>. The output, a significance map, is produced by doing a one-dimensional test for all possible resolution/position pairs. The significance map shows for which resolution/position pairs the null hypothesis is rejected. For the testing of multinormality, the Anderson-Darling test is utilized to detect potential departures from multinormality at different combinations of resolutions and positions. In the <i>k</i>-sample case, it is tested whether <i>k</i> data sets can be said to originate from the same unspecified discrete or continuous multivariate distribution. This is done by testing the <i>k</i> vectors corresponding to the same resolution/position pair of the k different data sets through the <i>k</i>-sample Anderson-Darling test. Successful demonstrations of the new methodology on artificial and real data sets are presented, and a feature selection scheme is demonstrated.en_US
dc.description.sponsorshipNational Science Foundation of the United Statesen_US
dc.descriptionSource at <a href=https://doi.org/10.1371/journal.pone.0211044>https://doi.org/10.1371/journal.pone.0211044</a>.en_US
dc.identifier.citationHindberg, K., Hannig, J. & Godtliebsen, F. (2019). A novel scale-space approach for multinormality testing and the k-sample problem in the high dimension low sample size scenario. <i>PLoS ONE, 14</i>(1), e0211044. https://doi.org/10.1371/journal.pone.0211044en_US
dc.identifier.cristinIDFRIDAID 1694524
dc.identifier.doi10.1371/journal.pone.0211044
dc.identifier.issn1932-6203
dc.identifier.urihttps://hdl.handle.net/10037/16390
dc.language.isoengen_US
dc.publisherPLOSen_US
dc.relation.journalPLoS ONE
dc.relation.projectIDinfo:eu-repo/grantAgreement/RCN/EVITA/176872/Norway/The Power of Scale-Space Methods and Gaussian Markov Random Fields Applied in Climatology and Medicine//en_US
dc.rights.accessRightsopenAccessen_US
dc.subjectVDP::Mathematics and natural science: 400::Mathematics: 410::Statistics: 412en_US
dc.subjectVDP::Matematikk og Naturvitenskap: 400::Matematikk: 410::Statistikk: 412en_US
dc.titleA novel scale-space approach for multinormality testing and the k-sample problem in the high dimension low sample size scenarioen_US
dc.typeJournal articleen_US
dc.typeTidsskriftartikkelen_US
dc.typePeer revieweden_US


File(s) in this item

Thumbnail

This item appears in the following collection(s)

Show simple item record