van Eggermond, Michael A.B.
Axhausen, Kay W.
Rights / licenseIn Copyright - Non-Commercial Use Permitted
Overcoming the “independence of irrelevant alternatives” (IIA) property of the basic multinomial logit (MNL) model is a major research issue in the field of discrete choice modelling. In recent years, several approaches have been developed to achieve this goal. This paper focuses on similarities in the context of air connection choice. Due to the extensive choice sets in this study a similarity factor approach is applied, which does - even for this huge data set - not prolong estimation time. The scrutinized similarity factor has been specifically developed for public transport connections. It is capable to account for the complex and multi-dimensional nature of similarities of public transport alternatives. The optimal parameter setting for the similarity factor with respect to the dataset at hand is found. The estimation results emphasize that the similarity between alternatives exerts a significant influence on air connection choice. Similarity is perceived positive. It appears that decision-makers have a strong preference for certain attributes of the alternatives such as a specific departure time, travel time or fare. Thus, they prefer alternatives that are similar with respect to these attributes. Show more
External linksSearch via SFX
Journal / seriesArbeitsberichte Verkehrs- und Raumplanung
PublisherETH, Eidgenössische Technische Hochschule Zürich, IVT
SubjectAIR TRAFFIC + AVIATION (TRANSPORTATION AND TRAFFIC); PASSENGER TRANSPORT (TRANSPORTATION AND TRAFFIC); TRAFFIC ANALYSIS (TRANSPORTATION AND TRAFFIC); PERSONENVERKEHR (VERKEHR UND TRANSPORT); LUFTVERKEHR (VERKEHR UND TRANSPORT); VERKEHRSANALYSE (VERKEHR UND TRANSPORT)
Organisational unit02610 - Inst. f. Verkehrspl. u. Transportsyst. / Inst. Transport Planning and Systems
03521 - Axhausen, Kay W. / Axhausen, Kay W.
02226 - NSL - Netzwerk Stadt und Landschaft / NSL - Network City and Landscape
02655 - Netzwerk Stadt und Landschaft D-ARCH
MoreShow all metadata