Outlier Detection Using Principle Components
by
Nedret Billor
Department of Mathematics and Statistics, Auburn University
Coauthors: Gulsen Kiral and Asuman Turkmen

Multivariate outliers are observations that are a long way from the rest of the observations in the p-dimensional space defined by the variables. There are several approaches for the detection of multivariate outliers. In this study we will consider the approaches to the detection of multivariate outliers using principal components. We examine the techniques for classically and robustly estimating principal components which are used to detect outliers and give the advantages and disadvantages of these techniques.

Date received: October 24, 2005

Copyright © 2005 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # carr-11.