CATEGORY ANALYSIS AND HOUSEHOLD MULTIPLE REGRESSION MODELS OF TRIP GENERATION: A POSSIBLE RECONCILLIATION

This paper shows that the main arguments employed in the debate concerning the respective advantages of category analysis and multiple regression models as techniques for forecasting trip generation and car ownership levels are incorrect and sometimes irrelevant. The author demonstrates that category analysis is nothing more than a specification of an all dummy variable multiple regression model. The creation of so-called "hybrid models" as a consequence of a combination of the category analysis approach with multiple regression gives more logical and reliable results. Such a synthesis, which follows after a number of conventional statistical tests to ensure the selection of the most appropriate specification, provides - according to the author - a much more realistic framework within which to incorporate the various influences on the trip generation and car ownership levels. /Author/

• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/03918440
• Authors:
  • BUTTON, K J
• Publication Date: 1976-12

Media Info

• Pagination: p. 19-27
• Serial:
  • International Journal of Transport Economics
  • Volume: 3
  • Issue Number: 3
  • ISSN: 0391-8440
  • Serial URL: http://www.libraweb.net/riviste.php?chiave=67

Subject/Index Terms

• TRT Terms: Analysis; Forecasting; Mathematical models; Motor vehicles; Multiple regression analysis; Ownership; Specifications; Statistics; Trip generation
• Uncontrolled Terms: Models
• Old TRIS Terms: Multiple regression
• Subject Areas: Data and Information Technology; Highways; Operations and Traffic Management;

Filing Info

• Accession Number: 00155974
• Record Type: Publication
• Report/Paper Numbers: Analytic
• Files: TRIS
• Created Date: Aug 4 1977 12:00AM