TRAVEL DEMAND FORECASTING BY USING THE NESTED MULTINOMIAL LOGIT MODEL

A considerable amount of recent travel demand research has highlighted the limiting assumptions of the multinomial logit model, particularly its property of being independent of irrelevant alternatives. Nevertheless, because of its tractability, the multinomial logit formulation is likely to remain the most important of disaggregate-analysis techniques. This paper points out that, although the axiom of the independence of irrelevant alternatives is a property of the simple multinomial logit model, a generalization of that model has been developed that is virtually free from that limitation, has been shown to be effective and economically usable in practical studies, and provides a simple diagnostic capability for assessing the validity of the assumption of independence of irrelevant alternatives in any given situation. The generalized logit model-referred to here as the nested multinomial logit model-has been reported in the literature for several years, but awareness of its properties and even existence seems to be very slight. This paper provides a background on the development of the nested multinomial logit model, presents its structure (with guidelines for its use), and reports on current research that uses the nested formulation as the analysis tool.

Media Info

  • Media Type: Print
  • Features: Figures; References; Tables;
  • Pagination: pp 48-55
  • Monograph Title: Travel demand models: application, limitations and quantitative methods
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00334196
  • Record Type: Publication
  • ISBN: 0309031192
  • Files: TRIS, TRB
  • Created Date: Oct 28 1981 12:00AM