Data from home-interview surveys in Detroit for 1953 and 1965 were used to test the time stability of disaggregate trip-generation and predistribution modal-choice models. Initial cross-classification analysis showed 4 to 18 percent increases in household trip-generation rates for households with cars available. A statistical test of the overall time stability of multiple linear regression trip-generation equations indicated that the equations were not stable unless non-trip-making households were removed. The individual regression coefficients also were tested for time stability, and, despite the lack of overall statistical time stability, disaggregate work- and home-based trip-generation equations for 1953 produced reasonable estimates of 1965 zone-level trips. Disaggregate regression equations for the automobile-driver and bus modes also were found to exhibit only limited time stability. Interaction between cars available and number of persons employed was particularly important in explaining bus trips. Tests of the time-stability assumption at the zone level were limited by the lack of zone-level interaction variables.

Media Info

  • Media Type: Print
  • Features: Figures; References; Tables;
  • Pagination: pp 76-86
  • Monograph Title: New approaches to travel forecasting
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00141292
  • Record Type: Publication
  • ISBN: 030902482X
  • Files: TRIS, TRB
  • Created Date: Oct 26 1976 12:00AM