Multivariate Outlier Detection With High-Breakdown Estimators
نویسندگان
چکیده
Multivariate Outlier Detection With High-Breakdown Estimators Andrea Cerioli Andrea Cerioli is Professor, Dipartimento di Economia, Sezione di Statistica e Informatica, Università di Parma, Via Kennedy 6, 43100 Parma, Italy . The author expresses his gratitude to three anonymous reviewers for insightful comments that led to many improvements in the article. The author also thanks Marco Riani and Anthony C. Atkinson for helpful discussions on previous drafts of this work.
منابع مشابه
The Largest Nonidentiiable Outlier: a Comparison of Multivariate Simultaneous Outlier Identiication Rules
The aim of detecting outliers in a multivariate sample can be pursued in diierent ways. We investigate here the performance of several simultaneous multivariate outlier identiication rules based on robust estimators of location and scale. It has been shown that the use of estimators with high nite-sample breakdown point in such procedures yields a good behaviour with respect to the prevention o...
متن کاملThe largest nonidentifiable outlier: A comparison of multivariate simultaneous outlier identification rules
The aim of detecting outliers in a multivariate sample can be pursued in di erent ways We investigate here the performance of several simultaneous multivariate outlier identi cation rules based on robust estimators of location and scale It has been shown that the use of estimators with high nite sample breakdown point in such procedures yields a good behaviour with respect to the prevention of ...
متن کاملThe minimum weighted covariance determinant estimator
In this paper we introduce weighted estimators of the location and dispersion of a multivariate data set with weights based on the ranks of the Mahalanobis distances. We discuss some properties of the estimators like the breakdown point, influence function and asymptotic variance. The outlier detection capacities of different weight functions are compared. A simulation study is given to investi...
متن کاملThe Maximum Asymptotic Bias of Outlier Identi
In their paper, Davies and Gather (1993) formalized the task of outlier identiica-tion, considering also certain performance criteria for outlier identiiers. One of those criteria, the maximum asymptotic bias, is carried over here to multivariate outlier identiiers. We show how this term depends on the respective biases of estimators which are used to construct the identiier. It turns out that ...
متن کاملThe masking breakdown point of multivariate outlier identification rules
In this paper we consider one step outlier identi cation rules for multivariate data generalizing the concept of so called outlier identi ers as presented in Davies and Gather for the case of univariate samples We investigate how the nite sample breakdown points of estimators used in these identi cation rules in uence the masking behaviour of the rules
متن کامل