Duopoly Strategies Programmed by Experienced Players
让经验丰富的受试者为20期非对称古诺寡头博弈编程策略，发现典型策略不依赖预期或优化，而是基于公平标准设定合作目标，并采用对等回应策略。
The strategy method asks experienced subjects to program strategies for a game. This paper reports on an application to a 20-period supergame of an asymmetric Cournot duopoly. The final strategies after three programming rounds show a typical structure. Typically, no expectations are formed and nothing is optimized. Instead of this, fairness criteria are used to determine cooperative goals, called ideal points. The subjects try to achieve cooperation by a measure-for-measure policy, which reciprocates movements towards and away from the ideal point by similar movements. A strategy tends to be more successful the more typical it is.