Equilibrium Selection in the Repeated Prisoner's Dilemma: Axiomatic Approach and Experimental Evidence
针对重复囚徒困境，提出一个公理化的均衡选择标准，该标准产生一个大于传统临界折现因子的新临界值，并在实验中展现出更强的预测能力。
We propose an axiomatic approach for equilibrium selection in the discounted, infinitely repeated symmetric Prisoner's Dilemma. Our axioms characterize a unique selection criterion that is also useful as a tool for applied comparative statics exercises as it results in a critical discount factor δ* strictly larger than δ, the standard criterion that has often been used in applications. In an experimental test we find a strong predictive power of our proposed criterion. For parameter changes where the standard and our criterion predict differently, changes in observed cooperation follow predictions based on δ*.