Commentary on "Analysis of Choice Expectations in Incomplete Scenarios"

This paper addresses an important topic in the preference elicitation field, namely the method of inferring population preference from hypothetical choice data collected from a random sample of that population. While criticized by a subset of economists who object to the hypothetical nature of the responses and put faith only in preference predictions derived from revealed preference analysis, hypothetical choice analysis is widely used in a range of economic applications that draw their data from, for example, panel studies. Manski attempts to provide a solid theoretical basis for the use of such response data. The focus of his present paper is the use of probabilistic expectations data to predict behavior in feasible or counterfactual scenarios. Given the rich context in which actual decisions are made, any choice scenario described by a researcher is bound to be incomplete. The contribution of the paper lies in its attempt to combine assumptions about respondents' preference formation with assumptions about their expectations of unspecified scenario details that may affect their preferences.