Robust versus classical detection of atypical data

Antille, Gérard; Ritschard, Gilbert

doi:10.1016/0167-9473(92)90082-Q

Scientific article

English

Robust versus classical detection of atypical data

ContributorsAntille, Gérard; Ritschard, Gilbert

Published inComputational statistics & data analysis, vol. 14, no. 1, p. 75-88

Publication date1992

Abstract

Residuals from a robust fit and robust distances are discussed and used as indicators of outliers or leverage points. The performance of these robust diagnostics are compared to the classical ones by means of a simulation study.

Affiliation entities

Faculté des sciences de la société / Institut de démographie et de socioéconomie

Citation (ISO format)

ANTILLE, Gérard, RITSCHARD, Gilbert. Robust versus classical detection of atypical data. In: Computational statistics & data analysis, 1992, vol. 14, n° 1, p. 75–88. doi: 10.1016/0167-9473(92)90082-Q

Article (Published version)

Identifiers

PID : unige:3743
DOI : 10.1016/0167-9473(92)90082-Q

Journal ISSN0167-9473

599views

777downloads

Creation06/10/2009 11:59:12

First validation06/10/2009 11:59:12

Update time14/03/2023 15:16:13

Status update14/03/2023 15:16:13

Last indexation29/10/2024 12:25:25

Archive ouverte UNIGE

Robust versus classical detection of atypical data

Residuals from a robust fit and robust distances are discussed and used as indicators of outliers or leverage points. The performance of these robust diagnostics are compared to the classical ones by means of a simulation study.

Technical informations