MICRO/THEORY: Confidence in Inference; Dr En Hua Hu (University of Oxford)

Abstract:

It is a commonly assumed that agents do not face uncertainty regarding the Blackwell experiment. I axiomatically characterize the empirical content of this assumption for an agent choosing between objects that are each associated with a sample of signals. The characterization does not rely on expected utility nor Bayesian updating. A simple thought experiment yields a natural choice pattern that these models cannot accommodate. In particular, the effect of increasing sample size on choice cannot be rationalized by these models. In a controlled experiment, 95% of subjects' choices violate models of certainty in information. Using a novel incentive-compatible confidence elicitation mechanism, I find confidence in correctly interpreting samples influences choice. As suggested by the thought experiment, and consistent with a model of uncertainty in information, many subjects display a sample size neglect bias which is positively associated with higher confidence.