Statistics Colloquium
April 4 (Fri) 11 a.m.
Alavi Commons Room, 6625 Everett Tower

A Cluster-Based Outlier Detection Scheme for Multivariate Data

J. Marcus Jobe
Professor of Information Systems and Analytics
Farmer School of Business
Miami University (Ohio)
Michael Pokojovy
University of Konstanz

Detection power of the squared Mahalanobis distance statistic is significantly reduced when several outliers exist within a multivariate data set of interest. To overcome this masking effect, we propose a computer-intensive cluster-based approach that incorporates a reweighted version of Rousseeuw’s minimum covariance determinant method with a multi-step cluster-based algorithm that initially filters out potential masking points. Compared to the most robust procedures, simulation studies show that our new method is better for outlier detection. Additional real data comparisons are given.

All statistics students are expected to attend.


Past colloquiums


Department of Statistics
3304 Everett Tower
Western Michigan University
Kalamazoo MI 49008-5152 USA
(269) 387-1420 | (269) 387-1419 Fax