Design Improved Parimutuel-type Information Aggregation Mechanisms: Inaccuracies and the Long-Shot Bias as Disequilibrium Phenomena

Abstract

Information Aggregation Mechanisms (IAMs) based on parimutuel-type betting systems can aggregate information from complex environments. However, the performance of previously studied systems leaves something to be desired due to possible bluffing, strategic timing of decisions and a so called long shot bias. This paper demonstrates that two modifications of parimutuel systems improve information aggregation performance by removing disinformation due to strategic behavior and by removing misleading disequilibrium behavior. The experiments also demonstrate that the so called long shot bias results from disequilibrium behavior as opposed to having roots in the psychology of the individuals.