Return to search

Multiple Outlier Detection: Hypothesis Tests versus Model Selection by Information Criteria

The detection of multiple outliers can be interpreted as a model selection problem. Models that can be selected are the null model, which indicates an outlier free set of observations, or a class of alternative models, which contain a set of additional bias parameters. A common way to select the right model is by using a statistical hypothesis test. In geodesy data snooping is most popular. Another approach arises from information theory. Here, the Akaike information criterion (AIC) is used to select an appropriate model for a given set of observations. The AIC is based on the Kullback-Leibler divergence, which describes the discrepancy between the model candidates. Both approaches are discussed and applied to test problems: the fitting of a straight line and a geodetic network. Some relationships between data snooping and information criteria are discussed. When compared, it turns out that the information criteria approach is more simple and elegant. Along with AIC there are many alternative information criteria for selecting different outliers, and it is not clear which one is optimal.

Identiferoai:union.ndltd.org:DRESDEN/oai:qucosa.de:bsz:520-qucosa-225770
Date14 June 2017
CreatorsLehmann, Rüdiger, Lösler, Michael
ContributorsHTW Dresden, Fakultät Geoinformation
PublisherHochschule für Technik und Wirtschaft Dresden
Source SetsHochschulschriftenserver (HSSS) der SLUB Dresden
LanguageEnglish
Detected LanguageEnglish
Typedoc-type:article
Formatapplication/pdf
SourceJournal of Surveying Engineering, Volume 142, Issue 4 (2016)

Page generated in 0.0027 seconds