Multinomial probit models are appealing for trip mode choice because they offer a flexible pattern of error correlation structure. This paper demonstrates the use of a multinomial probit model for work trip mode choice in Seoul, Korea. The Bayesian approach with Gibbs sampling is used. This method constructs a Markov chain Gibbs sampler that can be used to draw directly from the exact posterior distribution and perform finite sample likelihood inference. The authors estimate direct and cross-elasticities with respect to travel cost and the value of time. Findings indicate that travel demands are insensitive to cost but are sensitive to travel time change. The cross-elasticity results show that an increase in the cost of an automobile will increase the demand for bus transport more so than that of the subway. It is also shown that the bus has a greater substitution relative to the subway than the automobile (and vice versa). These results show that higher gains are more likely to come as a result of decreasing travel time for public transportation rather than by making it less expensive.

  • Availability:
  • Corporate Authors:

    Kluwer Academic Publishers

    P.O. Box 17
    Dordrecht,   Netherlands 
  • Authors:
    • Kim, Y
    • Kim, T-Y
    • Heo, E
  • Publication Date: 2003-8


  • English

Media Info

  • Features: References; Tables;
  • Pagination: p. 351-365
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00960730
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 1 2003 12:00AM