MODELLING SOUTH AUSTRALIAN ROAD CRASH CASUALTIES AND FATALITIES

This paper develops time series models to forecast South Australian monthly road crash casualties and fatalities using data from the South Australian Department of Transport road crash database. Data for the period January 1968 to June 1994 are analysed using a class of time series models known as the Autoregressive Integrated Moving Average (ARIMA) models. The approach developed by Box and Jenkins (1976) is adopted to statistically model crash casualties and fatalities on the basis of their historical patterns and forecast two years ahead to June 1996. The best ARIMA model for the casualty series is ARIMA (0,1,1)(0,1,1) while the model for the fatality series is ARIMA (0,1,1). The variability (standard errors of monthly estimates) for the fatality series was found to be high, generally in the order of 40 percent. It is concluded that these models are efficient for monitoring SA road crash casualties and fatalities. (a)

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/isbn/0730807533
• Corporate Authors:

  SOUTH AUSTRALIA. DEPARTMENT OF TRANSPORT. OFFICE OF ROAD SAFETY

  PO BOX 1
  WALKERVILLE, SOUTH AUSTRALIA  Australia  5081
• Authors:
  • BLACK-TIONG, P A
  • TIONG, F C
• Publication Date: 1996

Language

• English

Media Info

• Features: References;
• Pagination: 27 p.
• Serial:
  • Issue Number: 1/96

Subject/Index Terms

• TRT Terms: Crashes; Fatalities; Mathematical models; Methodology; Statistics
• Geographic Terms: Australia
• ITRD Terms: 1643: Accident; 8006: Australia; 1602: Fatality; 6473: Mathematical model; 9102: Method; 6555: Statistics
• Subject Areas: Data and Information Technology; Safety and Human Factors;

Filing Info

• Accession Number: 00731365
• Record Type: Publication
• Source Agency: ARRB
• ISBN: 0-7308-0753-3
• Files: ITRD, ATRI
• Created Date: Feb 27 1997 12:00AM