A generalized trade reduction mechanism
摘要
Double auction mechanism design centers on four properties: individual rationality (IR), incentive compatibility (IC), ideal efficiency, and budget balance (BB). The Myerson-Satterthwaite impossibility theorem prohibits satisfying all four properties simultaneously, but McAfee’s Trade Reduction Mechanism circumvents this by eliminating the least efficient trade. This article introduces a novel mechanism that optimizes the trade-off between efficiency and revenue in asymmetric markets. This control theoretic approach to mechanism design takes concepts drawn from George Leitmann’s adaptive stabilization strategies to adapt auction rules based on real-time feedback. The new mechanism proposed in this article generalizes trade reduction mechanisms through two key modifications: (1) the introduction of hyperparameters to establish distinct buyer and seller prices, ensuring positive revenue while maintaining efficiency, and (2) a "pre-filter" that detects inefficient traders, making the mechanism more resilient to market saturation. The resulting generalized trade reduction mechanism retains the economic properties of individual rationality, incentive compatibility, ideal efficiency, and budget balance.