María José Boccardi Chalela
Job Market Paper Title:
Predictive Ability and the Fit-Power Trade-Off in Theories of Consumption Behavior
Job Market Paper:
Utility maximization is a core assumption in economics for which the generalized axiom of revealed preference (GARP) provides a nonparametric test. Two problems arise when testing GARP. First, it provides an extremely sharp test since rejection occurs after only one violation. Second, even when a data set passes the test, it cannot necessarily be understood as a success for the theory since conditions may be so undemanding that any behavior would pass it. In this paper I propose a predictive ability approach to gauge the performance of the model. This approach provides a meaningful answer to the tension between the severity of the violations (fit) and the sensitivity of the test to detect them (power), a long standing problem in the literature. I present a nonparametric construction for the predictive distribution that can be used to forecast consumer's responses to policy or market interventions and to compute the marginal likelihood of the model. I summarize the informativeness of the predictive distribution using a set of intuitive statistics that yield a complete order across data sets for a given model. Additionally, I show how this approach can be used for model selection. I study the performance of these measures for the experimental data from Choi et al. (2007). As expected, the predictive precision measures are positively correlated with measures of fit and with the amount of revealed preference information that can be learned from data. I also find that more observations enhance predictive precision, and that GARP delivers a more precise predictive distribution than a model that assumes Cobb-Douglas preferences.
Behavioral Economics, Decision Theory, Economic Theory, Applied Econometrics