Controlling the size of multivariate outlier tests with the MCD estimator of scatter

Authors:
Andrea Cerioli;Marco Riani;Anthony C. Atkinson
Affiliations:
Dipartimento di Economia, Università di Parma, Parma, Italy 43100;Dipartimento di Economia, Università di Parma, Parma, Italy 43100;Department of Statistics, The London School of Economics, London, UK WC2A 2AE
Venue:
Statistics and Computing
Year:
2009

Citing 6
Cited 2

Robust regression and outlier detection

Robust regression and outlier detection
Mersenne twister: a 623-dimensionally equidistributed uniform pseudo-random number generator

ACM Transactions on Modeling and Computer Simulation (TOMACS) - Special issue on uniform random number generation
A fast algorithm for the minimum covariance determinant estimator

Technometrics
Influence function and efficiency of the minimum covariance determinant scatter matrix estimator

Journal of Multivariate Analysis
The largest nonindentifiable outlier: a comparison of multivariate simultaneous outlier identification rules

Computational Statistics & Data Analysis
MDQC

Bioinformatics

Error rates for multivariate outlier detection

Computational Statistics & Data Analysis
Detection of multivariate outliers in business survey data with incomplete information

Advances in Data Analysis and Classification

Quantified Score

Hi-index	0.00

Visualization

Abstract

Multivariate outlier detection requires computation of robust distances to be compared with appropriate cut-off points. In this paper we propose a new calibration method for obtaining reliable cut-off points of distances derived from the MCD estimator of scatter. These cut-off points are based on a more accurate estimate of the extreme tail of the distribution of robust distances. We show that our procedure gives reliable tests of outlyingness in almost all situations of practical interest, provided that the sample size is not much smaller than 50. Therefore, it is a considerable improvement over all the available MCD procedures, which are unable to provide good control over the size of multiple outlier tests for the data structures considered in this paper.