ON THE FRONTIER OF GENERATING REVEALED PREFERENCE CHOICE SETS: AN EFFICIENT APPROACH

Deterministic rules for generating choice sets are often employed by analysts confronting universal sets with large numbers of alternatives. For destination choice analysis, site exclusion rules defined by travel time, distance, or quality have a behavioral appeal, yet are fundamentally limited by their one-dimension scope. To remedy this shortcoming while maintaining the concept that trips require costly inputs to yield utility generating outputs, we develop and test an exclusion rule for generating choice sets defined by efficiency measures derived from stochastic frontier econometric models. Choice set composition, site choice efficiency and probability of selection, and consumer surplus are compared with results obtained under alternative exclusion rules.