Robust projection indices

GP Nason

Research output: Contribution to journalArticle (Academic Journal)peer-review

17 Citations (Scopus)

Abstract

Loosely speaking a robust projection index is one that prefers projections involving true clusters over projections consisting of a cluster and an outlier. We introduce a mathematical definition of one-dimensional index robustness and describe a numerical experiment to measure it. We design five new indices based on measuring divergence from Student's t-distribution which are intended to be especially robust: the experiment shows that they are more robust than several established indices. The experiment also reveals more generally that the robustness of moment indices depends on the number of approximation terms, providing additional practical guidance for existing projection pursuit implementations. We investigate the theoretical properties of one new Student t-index and Hall's index and show that the new index automatically adapts its robustness to the degree of outlier contamination. We conclude by outlining the possibilities for extending our experiments to both higher dimensions and other new indices.
Translated title of the contributionRobust projection indices
Original languageEnglish
Pages (from-to)551 - 567
Number of pages17
JournalJournal of the Royal Statistical Society Series B - Statistical Methodology
Volume63 (3)
DOIs
Publication statusPublished - Aug 2001

Bibliographical note

Publisher: Blackwell Publ Ltd

Fingerprint Dive into the research topics of 'Robust projection indices'. Together they form a unique fingerprint.

Cite this