EMPIRICAL APPROACHES TO OUTLIER DETECTION IN INTELLIGENT TRANSPORTATION SYSTEMS DATA
Novel methods for implementation of detector-level multivariate screening methods are presented. The methods use present data and classify data as outliers on the basis of comparisons with empirical cutoff points derived from extensive archived data rather than from standard statistical tables. In addition, while many of the ideas of the classical Hotelling's T-squared-statistic are used, modern statistical trend removal and blocking are incorporated. The methods are applied to intelligent transportation system data from San Antonio and Austin, Texas. These examples show how the suggested new methods perform with high-quality traffic data and apparently lower-quality traffic data. All algorithms were implemented by using the SAS programming language.
- Record URL:
- Summary URL:
-
Availability:
- Find a library where document is available. Order URL: http://www.trb.org/Main/Public/Blurbs/154629.aspx
-
Supplemental Notes:
- This paper appears in Transportation Research Record No. 1840, Statistical Methods and Modeling and Safety Data, Analysis, and Evaluation.
-
Corporate Authors:
500 Fifth Street, NW
Washington, DC United States 20001 -
Authors:
- Park, E S
- Turner, S
- Spiegelman, C H
- Publication Date: 2003
Language
- English
Media Info
- Features: Figures; References; Tables;
- Pagination: p. 21-30
-
Serial:
- Transportation Research Record
- Issue Number: 1840
- Publisher: Transportation Research Board
- ISSN: 0361-1981
Subject/Index Terms
- TRT Terms: Algorithms; Computer programming languages; Data banks; Intelligent transportation systems; Traffic data
- Uncontrolled Terms: Outliers
- Geographic Terms: Austin (Texas); San Antonio (Texas)
- Subject Areas: Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning;
Filing Info
- Accession Number: 00966622
- Record Type: Publication
- ISBN: 0309085810
- Files: TRIS, TRB, ATRI
- Created Date: Dec 16 2003 12:00AM