Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

If You Use a Screen Reader

This content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.

Assessment of Athens's Metro Passenger Behaviour via a Multiranked Probit Model

Michalis Linardakis and Petros Dellaportas
Journal of the Royal Statistical Society. Series C (Applied Statistics)
Vol. 52, No. 2 (2003), pp. 185-200
Published by: Wiley for the Royal Statistical Society
Stable URL: http://www.jstor.org/stable/3592703
Page Count: 16
  • Read Online (Free)
  • Download ($29.00)
  • Subscribe ($19.50)
  • Cite this Item
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Assessment of Athens's Metro Passenger Behaviour via a Multiranked Probit Model
Preview not available

Abstract

We deal with real data from a stated preference experiment which was designed to explain and predict passengers' behaviour towards three main means of transportation in the city of Athens. The resulting model formulations give rise to the so-called multiranked probit model which emerges from a series of ranking responses in a set of hypothetical scenarios, i.e. we enhance the multinomial probit model with the embodiment of a utility threshold parameter which deals realistically with ranking responses, intransitivity of indifference between alternatives or ties. Moreover, we ensure identifiable parameters for the covariance matrix of the underlying utility vectors, we include a hierarchical step that models the unit-specific utility thresholds as exchangeably distributed and, finally, we permit the use of heavy-tailed distributions for the stochastic error term. Our proposed methodology is Bayesian and the implementation tool adopted is Markov chain Monte Carlo sampling. The posterior output consists of practical information such as travel characteristics (e.g. walking times and waiting times), expressed either in drachmas per hour or in minutes of in-vehicle time, and 95% credible intervals of the probability of choosing a particular mode of transportation. These are key factors in determining whether a policy has positive or negative net benefits.

Page Thumbnails

  • Thumbnail: Page 
[185]
    [185]
  • Thumbnail: Page 
186
    186
  • Thumbnail: Page 
187
    187
  • Thumbnail: Page 
188
    188
  • Thumbnail: Page 
189
    189
  • Thumbnail: Page 
190
    190
  • Thumbnail: Page 
191
    191
  • Thumbnail: Page 
192
    192
  • Thumbnail: Page 
193
    193
  • Thumbnail: Page 
194
    194
  • Thumbnail: Page 
195
    195
  • Thumbnail: Page 
196
    196
  • Thumbnail: Page 
197
    197
  • Thumbnail: Page 
198
    198
  • Thumbnail: Page 
199
    199
  • Thumbnail: Page 
200
    200