Transposition invariant principal component analysis in L1 for long tailed data

Similar to the ordinary principal component analysis (PCA), we develop PCA in L1 satisfying an invariance property: The objective function, which is a matrix norm, is transposition invariant. The new method is robust and specifically useful for long-tailed data. An example is provided.

MoreLess

Year of publication:	2005
Authors:	Choulakian, Vartan
Published in:	Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 71.2005, 1, p. 23-31
Publisher:	Elsevier
Keywords:	PCA Centroid method Transposition invariant matrix norms Transition formulae

More details

Type of publication:	Article
Source:	RePEc - Research Papers in Economics

Persistent link: https://www.econbiz.de/10005137843