AbstractIn many discrete choice contexts the actual choice set, including the alternatives effectively perceived and considered by the decision maker, may substantially differ from the universal choice set, including all available alternatives: one of the most relevant examples within transport demand simulation is probably the choice of destination, wherein the universal choice set normally includes hundreds of traffic zones. In these cases, proper simulation of the choice set is crucial for correct simulation of the choice context.In this regard, our paper has two main objectives. The first is to give a general contribution to choice set modelling by extending and applying the concept of dominance among alternatives to the framework of random utility theory. The main result is the definition of a methodology for the generation of new dominance attributes, which can be used in choice set modelling. The second aim is to make a specific contribution to destination choice modelling: dominance attributes are defined from the above methodology and introduced into this choice context, and new spatial variables reproducing better knowledge of zones with a privileged spatial position are also proposed. Methodology and attributes are tested both on synthetic and on real data.
Dominance among alternatives in random utility models
Transportation Research Part A: Policy and Practice ; 43 , 2 ; 170-179
2008-10-28
10 pages
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
Dominance among alternatives in random utility models
Online Contents | 2009
|Dominance among alternatives in random utility models
Online Contents | 2009
|British Library Conference Proceedings | 2005
|Sampling of Alternatives in Random Regret Minimization Models
Online Contents | 2016
|