Do We Follow Others when We Should? A Simple Test of Rational Expectations: Comment

摘要

Weizsaecker (2010)-henceforth, W-presents a meta-dataset from 13 information cascade experiments and an appealing reduced-form approach to measure the success of social learning in these experiments. The payoff of actions is estimated, and the resulting empirical payoff is controlled for in regression analyses to examine to what degree participants choose the empirically optimal action, i.e., to formulate a test of rational expectations about the underlying payoffs of actions. W concludes that participants perform poorly when trying to extract the information that is contained in previous actions and that rational expectations are violated in several systematic ways.