不定次重复的古诺模型博弈及其仿真研究

Research on Cournot model game with uncertain repeat times and its simulation

重复次数不确定的重复博弈比无限次重复博弈更符合客观实际. 提出了不定次重复的古诺模型及不同的博弈策略;通过计算机仿真的方法,根据设计的仿真方案对不定次重复的古诺模型博弈过程进行了仿真,得到了不同贴现因子条件下各种仿真方案的厂商得益;对各方案的结果进行了分析比较,得到了统计意义上的临界贴现因子和均衡策略;阐释了仿真结果的输入弹性的概念,并用其分析了不同仿真方案的结果差异的原因.

Repeated games with uncertain repeat times are more consistent with natural practice than repeated game with unlimited times. First, this paper proposes both the Cournot model with uncertain repeat times and different game strategies. Second, the paper simulates the course of Cournot model game with uncertain repeat times according to the designed simulation plans. The players' gains for different discount rates are obtained under the condition of each simulation plan. Finally, the paper analyzes the results of every simulation plan and works out the critical discount rate and equilibrium strategy in the statistical sense. In addition, this paper explains the concept of input elasticity of simulation results and analyses the reason why there are differences in the results of different simulation strategies using the concept.