A BAYESIAN ANALYSIS OF THE "ACCIDENT RATE POTENTIAL" PARAMETER

Previous work by others included 90 per cent confidence limits on the parameter lambda, an individual's accident rate potential derived from a given number of accidents in A given time period. However, due to wide and overlapping intervals, virtually no discrimination between individuals was possible. This paper presents a detailed statistical analysis of the accident data from the original work and describes the calculations for the parameter of the distribution of accident rate potential. It details the use of the bayes' theorem in deriving confidence intervals as percentage points of a posterior distribution of lambda such that discrimination between individuals with different accident records can be made. /TRRL/

  • Availability:
  • Corporate Authors:

    Pergamon Press, Incorporated

    Maxwell House, Fairview Park
    Elmsford, NY  United States  10523
  • Authors:
    • Williford, W O
    • Barton, H E
  • Publication Date: 1975-5

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00133701
  • Record Type: Publication
  • Source Agency: Transport and Road Research Laboratory (TRRL)
  • Files: ITRD, TRIS
  • Created Date: Sep 4 1976 12:00AM