Universidad de Montevideo, Facultad de Ciencias Empresariales y Economía, Departamento de Economía
Fecha
2013
Extensión
31 p.
Resumen
We conduct two experiments of the claim that people are overconfident, using newtests of overplacement that are based on a formal Bayesian model. Our two experi-ments, on easy quizzes, find overplacement. More precisely, we find apparently over-confident data that cannot be accounted for by a rational population of expected utilitymaximizers with a good understanding of the nature of the quizzes they took. The finding is of particular interest because Benoit and Dubra (2011) have shown that thevast majority of the existing findings on the better-than-average e§ect are actuallyconsistent with Bayesian updating.