Outlier Detection in the Multiple Cluster Setting Using the Minimum Covariance Determinant Estimator

Authors

Johanna S. Hardin, Pomona CollegeFollow
David M. Rocke, University of California - Davis

Document Type

Article

Department

Mathematics (Pomona)

Publication Date

2004

Keywords

Minimum covariance determinant, Robust clustering, Outlier detection

Abstract

Mahalanobis-type distances in which the shape matrix is derived from a consistent high-breakdown robust multivariate location and scale estimator can be used to find outlying points. Hardin and Rocke (http://www.cipic.ucdavis.edu/~dmrocke/preprints.html) developed a new method for identifying outliers in a one-cluster setting using an F distribution. We extend the method to the multiple cluster case which gives a robust clustering method in conjunction with an outlier identification method. We provide results of the Fdistribution method for multiple clusters which have different sizes and shapes.

Rights Information

© 2002 Elsevier B.V.

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

DOI

10.1016/S0167-9473(02)00280-3

Recommended Citation

Johanna Hardin, David M Rocke, Outlier detection in the multiple cluster setting using the minimum covariance determinant estimator, Computational Statistics & Data Analysis, Volume 44, Issue 4, 28 January 2004, Pages 625-638, ISSN 0167-9473, http://dx.doi.org/10.1016/S0167-9473(02)00280-3. (http://www.sciencedirect.com/science/article/pii/S0167947302002803)