MULTIDIMENSIONAL CHOICE MODELS: ALTERNATIVE STRUCTURES OF TRAVEL DEMAND MODELS

Choice theories in travel demand modeling assume a selection from a finite set of mutually exclusive and collectively exhaustive alternatives. It is assumed that, with qualitative or discrete alternatives, probabilistic behavior explains observations of different choices for the same set of observed independent variables. Such choice theories have been developed in the context of unidimensional choice situations. The total number of choices that a consumer makes is very large. The assumptions of a "utility tree," or a separable utility function, and negligible income effects permit the independent modeling of demand for a subset of commodities. Mobility and travel choices are such an independent branch or subset of the consumer's utility function. Choices within this subset are interdependent. This subset may be treated as a block recursive system. That is, the first block consists of the mobility choices, and the second block consists of the travel choices (assuming the mobility choices as fixed). Travel choices with respect to different trip purpose categories can also be considered independently of each other. Thus, one can model separately the set of mobility choices and the sets of travel choices for different trip purposes (assuming that mobility choices are predetermined). Yet, each of the above sets of choices represents a multidimensional choice situation. The purpose of this paper is to extend the choice theories from unidimensional to multidimensional choice situations. In a multidimensional choice situation different assumptions about the dependencies among choices result in models with different structures. The alternative structures are identified, and their applicability to travel demand models is discussed.

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00081605
  • Record Type: Publication
  • Files: TRIS, TRB, ATRI
  • Created Date: Apr 8 1975 12:00AM