THE LOCAL SPATIAL AUTOCORRELATION AND THE KERNEL METHOD FOR IDENTIFYING BLACK ZONES: A COMPARATIVE APPROACH

Two methos are presented and compared in this paper for identifying and delimiting road sections characterized by concentration of road accidents: one method based on spatial autocorrelation indices, the other one on kernel estimators. The paper shows that although the two methods start from a different conceptual approach, both may provide quite similar results under a specific choice of parameters and lead to a definition of non-contiguous black zones.

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/00014575
• Corporate Authors:

  Elsevier

  The Boulevard, Langford Lane
  Kidlington, Oxford  United Kingdom  OX5 1GB
• Authors:
  • Flahau, B
  • MOUCHART, M
  • SAN MARTIN, E
  • Thomas, I
• Publication Date: 2003-11

Language

• English

Media Info

• Features: Figures; References; Tables;
• Pagination: p. 991-1004
• Serial:
  • Accident Analysis & Prevention
  • Volume: 35
  • Issue Number: 6
  • Publisher: Elsevier
  • ISSN: 0001-4575
  • Serial URL: http://www.sciencedirect.com/science/journal/00014575

Subject/Index Terms

• TRT Terms: Autocorrelation; High risk locations; Traffic crashes
• Uncontrolled Terms: Black zones; Kernel estimation
• Subject Areas: Highways; Safety and Human Factors;

Filing Info

• Accession Number: 00963985
• Record Type: Publication
• Files: TRIS, ATRI
• Created Date: Oct 3 2003 12:00AM