Feature Informativeness, Curse-of-Dimensionality and Error Probability in Discriminant Analysis

Research output: ThesisDoctoral Thesis (compilation)

Abstract

This thesis is based on four papers on high-dimensional discriminant analysis. Throughout, the curse-of-dimensionality effect on the precision of the discrimination performance is emphasized. A growing dimension asymptotic approach is used for assessing this effect and the limiting error probability are taken as the performance criteria. A combined effect of a high dimensionality and feature informativeness on the discrimination performance is evaluated.

Details

Authors
  • Tatjana Pavlenko
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Probability Theory and Statistics

Keywords

  • Statistics, Matematik, Mathematics, feature selection, Discriminant analysis, feature informativeness, growing dimension assymptotics, operations research, operationsanalys, programmering, aktuariematematik, programming, actuarial mathematics, Statistik
Original languageEnglish
QualificationDoctor
Awarding Institution
Supervisors/Assistant supervisor
  • [unknown], [unknown], Supervisor, External person
Award date2001 Jun 6
Print ISBNs91-628-4775-9
Publication statusPublished - 2001
Publication categoryResearch

Bibliographic note

Defence details Date: 2001-06-06 Time: 10:15 Place: sal MH:C External reviewer(s) Name: Fang, K.T. Title: Professor Affiliation: [unknown] --- Pavlenko,~T. and von Rosen,~D.: Effect ofDimensionality on Discrimination. To appear in {it Statistics}, 2001. Pavlenko,~T.: Feature Informativeness inHigh-Dimensional Discriminant Analysis. Preliminary accepted by{it Communications in Statistics}, 2001.Pavlenko,~T.: On Assessing the FeatureInformativeness in High-Dimensional Discriminant Analysis.In {it Proceedings the Third Ukrainian-Scandinavian Conference inProbability Theory and Mathematical Statistics}, Kiev, Ukraine, June1999.Pavlenko,~T.: On Feature Selection,Curse-of-Dimensionality and Error Probability inDiscriminant Analysis. Submitted to {it Journal of StatisticalPlanning and Inference}.