RISK CURVE BOUNDARIES

This article provides an analysis of injury risk curves in terms of their usability and appropriateness with respect to: sample size, sensoring, shape of the underlying risk function, the inclusion of actual along with sensored data, and stimulus distribution over the critical range - the interval that spans from the highest stimulus value for which there is no injury to the lowest stimulus value for which there will always be an injury. The results indicate that generally the estimation error decreases as either the total sample size or the number of actual data increases; and, the more sharply increasing is the underlying risk function, the smaller is the sample size needed to obtain good risk estimations and the less influential is the availability of actual injury data.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/49192340
• Corporate Authors:

  Taylor & Francis

  4 Park Square, Milton Park
  Abingdon,   United Kingdom  OX14 4RN
• Authors:
  • DiDomenico, L
  • Nusholtz, G
• Publication Date: 2005

Language

• English

Media Info

• Features: Appendices; Figures; References; Tables;
• Pagination: p. 86-94
• Serial:
  • Traffic Injury Prevention
  • Volume: 6
  • Issue Number: 1
  • Publisher: Taylor & Francis
  • ISSN: 1538-9588
  • Serial URL: http://www.tandf.co.uk/journals/titles/15389588.html

Subject/Index Terms

• TRT Terms: Confidence intervals; Data collection; Data quality; Injury characteristics; Injury rates; Risk analysis; Statistical distributions
• Subject Areas: Highways; Safety and Human Factors;

Filing Info

• Accession Number: 00989581
• Record Type: Publication
• Files: TRIS
• Created Date: Apr 19 2005 12:00AM