Deductive versus inductive equilibrium selection: experimental results

The debate in equilibrium selection appears to have culminated in the formation of two schools of thought: those who favor equilibrium selection based on rational coordination and those who favor zero-rationality adaptation. We examine four deductive selection principles and find that each fails to explain experimental data. We propose an inductive selection principle based on simple learning dynamics. Using out-of-sample maximum likelihood parameters, the predictive performance of one such dynamic is shown to be dramatically better than the deductive selection principles. However, this selection principle is not always definitive, since no dynamic is guaranteed to converge. © 2003 Elsevier B.V. All rights reserved. JEL classification: B41; C14; C51; C72; C90