The Extent of Information: Its Relevance for Random Utility Models
This paper examines the effect of the researcher's definition of a recreator's choice set on the parameter and welfare estimates from standard random utility models. Three choice set definitions are compared: choice sets based on all alternatives within the survey, those alternatives within a specified distance from the respondents home, and alternatives familiar to the individual. We show that parameter and welfare estimates are sensitive to the definition of the choice set. Welfare measures of site closures differ in varying ways depending on the familiarity level of the sites in question.