Robust Automatic Methods for Outlier and Error Detection
Robust Automatic Methods for Outlier and Error Detection
Editing in surveys of economic populations is often complicated by the fact that outliers due to errors in the data are mixed in with correct, but extreme, data values. In this paper we describe and evaluate two automatic techniques for error identification in such long tailed data distributions. The first is a forward search procedure based on finding a sequence of error-free subsets of the error contaminated data and then using regression modelling within these subsets to identify errors. The second uses a robust regression tree modelling procedure to identify errors. Both approaches can be implemented on a univariate basis or on a multivariate basis. An application to a business survey data set that contains a mix of extreme errors and true outliers is
described.
Southampton Statistical Sciences Research Institute, University of Southampton
Chambers, Ray
96331700-f45e-4483-a887-fef921888ff2
Hentges, Adão
fe5f9c66-e8cf-40f5-83b7-c69d95277849
Zhao, Xinqiang
624c169e-a7fa-4a9a-bc78-a9c1b9d93992
2003
Chambers, Ray
96331700-f45e-4483-a887-fef921888ff2
Hentges, Adão
fe5f9c66-e8cf-40f5-83b7-c69d95277849
Zhao, Xinqiang
624c169e-a7fa-4a9a-bc78-a9c1b9d93992
Chambers, Ray, Hentges, Adão and Zhao, Xinqiang
(2003)
Robust Automatic Methods for Outlier and Error Detection
(S3RI Methodology Working Papers, M03/17)
Southampton, UK.
Southampton Statistical Sciences Research Institute, University of Southampton
29pp.
Record type:
Monograph
(Project Report)
Abstract
Editing in surveys of economic populations is often complicated by the fact that outliers due to errors in the data are mixed in with correct, but extreme, data values. In this paper we describe and evaluate two automatic techniques for error identification in such long tailed data distributions. The first is a forward search procedure based on finding a sequence of error-free subsets of the error contaminated data and then using regression modelling within these subsets to identify errors. The second uses a robust regression tree modelling procedure to identify errors. Both approaches can be implemented on a univariate basis or on a multivariate basis. An application to a business survey data set that contains a mix of extreme errors and true outliers is
described.
More information
Published date: 2003
Identifiers
Local EPrints ID: 8167
URI: http://eprints.soton.ac.uk/id/eprint/8167
PURE UUID: 4d17cb45-a9c1-4860-a3a4-4c1414319735
Catalogue record
Date deposited: 11 Jul 2004
Last modified: 15 Mar 2024 04:51
Export record
Contributors
Author:
Ray Chambers
Author:
Adão Hentges
Author:
Xinqiang Zhao
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics