STATISTICAL ESTIMATION OF INDIVIDUAL MOTOR VEHICLE DRIVER'S ACCIDENT LIABILITY BASED ON AN ANALYSIS OF THE MEAN TIME INTERVALS BETWEEN ACCIDENTS OF GEORGIA DRIVERS

The authors discuss the disadvantages encountered in the application of certain statistical techniques used in the accident analysis of groups of people, in the requirement that the accident data must be collected from a group of people who were exposed to risk for the same period of time. The derivation of statistical methods is discussed, which only involve data concerning a single individual and thus provide statistics from which influences centred on the individual can be drawn. In this paper the authors analyse and compare these and some other related statistical techniques using accident and citation records of individual Georgia motor vehicle drivers. The minimum number of accidents or citations associated with a single individual is also ascertained, in order to satisfy the required assumptions and achieve the required reasonable degree of statistical reliability. The data analysed were obtained from the department of public safety for the state of Georgia, and is described as a random sample of approx 2 1/2 per cent of the motor vehicle drivers licensed by the state of Georgia designed to obtain a group of drivers with the same exposure period. With a sample size of 46978 males and 33695 females, the information used in the analysis included for each individual driver age, sex, citation and accident incidence in Georgia covering a period of seven years. Fifty drivers were also analysed individually, chosen for their varied accident (one or more) and citation history. Details are provided of the mean time model and statistical estimation technique employed in the analysis, and the results of accident liability estimates and citation liability estimates presented in a series of tables. These results are discussed and evaluated. A Bayesian approach to the problem is presented and analysed and the results tabulated. A discussion follows on the assumptions and the estimators and their reliability. /TRRL/

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

  Pergamon Press, Incorporated

  Maxwell House, Fairview Park
  Elmsford, NY  United States  10523
• Authors:
  • Williford, W O
  • Murdock, G R
• Publication Date: 1978-9

Language

• English

Media Info

• Features: References; Tables;
• Pagination: p. 189-205
• Serial:
  • Accident Analysis & Prevention
  • Volume: 10
  • Issue Number: 3
  • Publisher: Elsevier
  • ISSN: 0001-4575
  • Serial URL: http://www.sciencedirect.com/science/journal/00014575

Subject/Index Terms

• TRT Terms: Age; Crash analysis; Crash rates; Crashes; Crimes; Driver records; Drivers; Females; Gender; Hazards; High risk drivers; Human beings; Mathematical models; Recidivists; Significance (Statistics); Statistical analysis; Statistical sampling; Statistics; Time; Traffic crashes; Traffic safety
• Uncontrolled Terms: Driver age; Violations
• Geographic Terms: United States
• ITRD Terms: 1643: Accident; 1612: Accident rate; 1757: Age; 1787: Man; 6473: Mathematical model; 1516: Offence; 1519: Recidivist; 9150: Risk; 6553: Sample (stat); 6532: Significance (stat); 6555: Statistics; 5414: Time; 8122: USA; 1788: Woman
• Subject Areas: Data and Information Technology; Highways; Safety and Human Factors; Security and Emergencies; Society; I81: Accident Statistics;

Filing Info

• Accession Number: 00189629
• Record Type: Publication
• Source Agency: Transport and Road Research Laboratory (TRRL)
• Files: ITRD, TRIS
• Created Date: May 11 1979 12:00AM