A new non-parametric detector of univariate outliers for distributions with unbounded support

Jean-Marc Bardet (1) , Solohaja Faniaha Dimby (1)

1 SAMM - Statistique, Analyse et Modélisation Multidisciplinaire (SAmos-Marin Mersenne)

The purpose of this paper is to construct a new non-parametric detector of univariate outliers and to study its asymptotic properties. This detector is based on a Hill's type statistic. It satisfies a unique asymptotic behavior for a large set of probability distributions with positive unbounded support (for instance: for the absolute value of Gaussian, Gamma, Weibull, Student or regular variations distributions). We have illustrated our results by numerical simulations which show the accuracy of this detector with respect to other usual univariate outlier detectors (Tukey, MAD or Local Outlier Factor detectors). The detection of outliers in a database providing the prices of used cars is also proposed as an application to real-life database.

Mots clés: Hill estimator; order statistics; non-parametric test; Outlier detection

Domaines: Statistiques [math.ST]

Jean-Marc Bardet, Solohaja Faniaha Dimby. A new non-parametric detector of univariate outliers for distributions with unbounded support. Extremes, 2017, 20 (4), pp.751-775. ⟨10.1007/s10687-017-0295-3⟩. ⟨hal-01196005v2⟩