Simple Models for Multi-Attribute Choice with Many Alternatives: When it Does and Does Not Pay to Face Tradeoffs with Binary Attributes
Universitat Pompeu Fabra Working Paper
Posted: 8 May 2004
Date Written: February 2004
Abstract
People experience difficulty in facing tradeoffs in multi-attribute choice. However, to what extent do they need to face these tradeoffs explicitly? We examine choices by simple tradeoff - avoiding models in environments where "true" preferences are linear and attributes are characterized by binary cues. We show that a deterministic elimination-by-aspects model (DEBA) is highly effective over a range of conditions. When "true" preferences are quite compensatory, however, a hybrid model that first chooses by equal weighting (EW) and then uses DEBA to resolve ties (if necessary) is more effective. We explore the sensitivity of these results to errors in the application of DEBA, to characteristics of distributions of alternatives, and to the presence of error in "true" preferences. We conclude that the outputs of these "boundedly rational" choice models are almost indistinguishable from those of "rational" models under the conditions examined here. Future research should explore how, if, and when these results can be extended to other circumstances.
Keywords: Decision making, bounded rationality, lexicographic rules, choice theory
JEL Classification: D81, M10
Suggested Citation: Suggested Citation