2010-05-11Diskussionspapier DOI: 10.18452/4254
Non-Gaussian Component Analysis
 dc.contributor.author Panov, Vladimir dc.date.accessioned 2017-06-16T00:08:27Z dc.date.available 2017-06-16T00:08:27Z dc.date.created 2010-05-27 dc.date.issued 2010-05-11 dc.identifier.issn 1860-5664 dc.identifier.uri http://edoc.hu-berlin.de/18452/4906 dc.description.abstract In this article, we present new ideas concerning Non-Gaussian Component Analysis (NGCA). We use the structural assumption that a high-dimensional random vector X can be represented as a sum of two components - a lowdimensional signal S and a noise component N. We show that this assumption enables us for a special representation for the density function of X. Similar facts are proven in original papers about NGCA ([1], [5], [13]), but our representation differs from the previous versions. The new form helps us to provide a strong theoretical support for the algorithm; moreover, it gives some ideas about new approaches in multidimensional statistical analysis. In this paper, we establish important results for the NGCA procedure using the new representation, and show benefits of our method. eng dc.language.iso eng dc.publisher Humboldt-Universität zu Berlin, Wirtschaftswissenschaftliche Fakultät dc.rights.uri http://rightsstatements.org/vocab/InC/1.0/ dc.subject dimension reduction eng dc.subject non-Gaussian components eng dc.subject EDR subspace eng dc.subject classification problem eng dc.subject Value at Risk eng dc.subject.ddc 330 Wirtschaft dc.title Non-Gaussian Component Analysis dc.type workingPaper dc.identifier.urn urn:nbn:de:kobv:11-100111393 dc.identifier.doi http://dx.doi.org/10.18452/4254 local.edoc.pages 23 local.edoc.type-name Diskussionspapier local.edoc.container-type series local.edoc.container-type-name Schriftenreihe local.edoc.container-year 2010 dc.title.subtitle New Ideas, New Proofs, New Applications dc.identifier.zdb 2195055-6 bua.series.name Sonderforschungsbereich 649: Ökonomisches Risiko bua.series.issuenumber 2010,26