A non-compensatory choice model incorporating attribute cutoffs |
| |
Authors: | Joffre Swait |
| |
Institution: | University of Florida and Advanis Inc., 12 West University Avenue, Suite 205, Gainesville, FL 32601, USA |
| |
Abstract: | This research proposes an extension to the traditional compensatory utility maximization framework which has guided most theoretical and statistical work in choice modeling applications, including those in transportation demand estimation work. Attribute cutoffs are incorporated into the decision problem formulation; it is then argued on extant empirical evidence that individuals may view these constraints as “soft”. This leads to the formulation of a penalized utility function that allows for constraint violation, but at a cost to the overall evaluation of the good. The proposed model is able to represent fully compensatory, conjunctive and disjunctive choice strategies, as well as combinations thereof. The properties of the proposed theoretical model are examined and discussed. From the theoretical framework, statistical models of choice behavior are easily derived; in their simplest forms, these models can be estimated using existing software. A Stated Preference choice experiment is analyzed using the proposed model, which is found to be highly consistent with observed choices and superior to a structural two-stage choice set formation model. |
| |
Keywords: | Non-compensatory choice Heuristic decision-making SP data |
本文献已被 ScienceDirect 等数据库收录! |
|