适应性预期、复杂网络与标准扩散动力学——基于计算经济学的研究  

Adaptive Expectation,Complex Network and the Dynamic of Standard Diffusion——Research Based on Computational Economics

作　　者：鲜于波[1] 梅琳[1] 

机构地区：[1]中山大学岭南学院,广州511275

出　　处：《管理科学》2007年第4期62-71,72,共11页Journal of Management Science

基　　金：国家自然科学基金(70573125);中山大学985工程产业与区域发展研究哲学社会科学创新基地资助项目

摘　　要：标准扩散是现代产业经济中的重要问题。主要采用复杂网络的小世界网络模型和基于agent的计算机建模技术,结合行为博弈学习理论中的EWA算法,对网络外部性下标准扩散过程中的适应性预期进行研究。基于个体在局部网络中的交互作用和个体行为的学习演化与预期自适应调整,通过计算机仿真就适应性预期对标准扩散的影响从多方面进行分析。研究结果揭示了个体适应性预期在标准扩散中的重要作用,发现扩散中的"涌现"现象,个体邻域结构的随机性对标准扩散结果及其最终分布状态有很大的影响,个体预期能够逆转标准扩散中的先发优势,预期速度调整因子在不同预期强度下有完全不同的影响,个体对新标准的较高预期也是新标准得以扩散的必要条件。Standard diffusion is an important problem in modern industry economy. This paper analyses the role of expectation in standard diffusion with the network externality using the theory of complex network from the point of agent-based computational modeling. At the same time, this paper also includes the results achieved in behavior game theory, especially the EWA algorithm. Based on the local interaction of agents and the new agent learning algorithms, the paper analyses the adaptive expectation and its influence in standard diffusion process. The simulation shows that agent expectation has great effect on standard diffusion process and expectation will change the situation of first move advantage. Results also show the ＂emergent＂ phenomena in standard diffusion process. The strength of expectation, the adjustment speed of expectation and agent neighborhoods all have obvious effect on standard diffusion process. Also, higher expectation is the necessary condition of new standard diffusion among consumers.

关 键 词：网络外部性 标准扩散 复杂网络 基于主体的计算经济学 适应性预期 

分 类 号：F062.9[经济管理—政治经济学]