This paper introduces a variant of random utility choice models based on mixed probability density functions, hence the adopted moniker "k-Mix models." Mixed pdf's contain components with the usual continuous density function specifications that underlie common choice models (e.g. MNL, GEV, MNP), but also contain one or more discrete probability mass points. These mixed pdf's result in models that can be interpreted to reflect different regimes of decision-making. Two exemplars developed in this paper, the 2- and 3-Mix models, are the result of a mixed pdf that combines a continuous pdf, plus one or two mass points, respectively. The 2-Mix permits a specific alternative to be in the Tradeoff Condition (the usual situation for alternatives in extant choice models, and the only regime in which compensatory utility is defined and compared) or in the Rejection Condition (in which an alternative has extreme disutility). The 3-Mix model adds the Dominance Condition (in which an alternative has an extremely attractive utility) - interestingly, the inclusion of this condition makes the model capable of simulating a particular form of satisficing decision making. These models are derived, discussed and compared relative to several extant choice models, then applied empirically on both a RP data set (work trip mode choice) and a SP data set (recreation campsite selection).
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
As the access to this document is restricted, you may want to look for a different version under "Related research" (further below) or search for a different version of it.
Volume (Year): 43 (2009) Issue (Month): 7 (August) Pages: 766-783 Download reference. The following formats are available: HTML
(with abstract),
plain text
(with abstract),
BibTeX,
RIS (EndNote, RefMan, ProCite),
ReDIF