EMPIRICAL APPROACHES TO OUTLIER DETECTION IN ITS DATA

This paper presents novel methods for implementing detector-level multivariate-screening methods. The methods use current day data and classify data as outliers based on comparisons to empirical cutoff points derived from extensive archived data rather than from standard statistical tables. In addition, while many of the ideas of classical Hotelling's T2 statistic are used, modern statistical trend removal and blocking are incorporated. The methods are applied to ITS data from San Antonio and Austin, Texas. These examples show how the suggested new methods perform on high quality traffic data and apparently lesser quality traffic data. All our algorithms were implemented using SAS

  • Supplemental Notes:
    • Publication Date: 2003. Transportation Research Board, Washington DC. Remarks: Paper prepared for presentation at the 82nd annual meeting of the Transportation Research Board, Washington, D.C., January 2003. Format: CD ROM
  • Corporate Authors:

    University of California, Berkeley

    California PATH Program, Institute of Transportation Studies
    Richmond Field Station, 1357 South 46th Street
    Richmond, CA  United States  94804-4648

    California Department of Transportation

    1120 N Street
    Sacramento, CA  United States  95814

    University of California, Berkeley

    Department of Electrical Engineering and Computer Sciences
    Berkeley, CA  United States  94720
  • Authors:
    • Park, Eun Sug
    • Turner, Shawn
    • Spiegelman, Clifford H
  • Conference:
  • Date: 2003

Language

  • English

Media Info

  • Pagination: 23 p.

Subject/Index Terms

Filing Info

  • Accession Number: 00941670
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: PATH, STATEDOT
  • Created Date: May 1 2003 12:00AM