主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2008, Vol. 16 ›› Issue (1): 172-179.

• 论文 • 上一篇    下一篇

基于EWA博弈学习模型的股权分置改革对价均衡研究

饶育蕾, 徐艳辉   

  1. 中南大学商学院 湖南长沙410083
  • 收稿日期:2007-06-26 修回日期:2007-12-23 出版日期:2008-02-28 发布日期:2008-02-28
  • 作者简介:饶育蕾(1964- ),女(汉族),四川人,中南大学商学院任职教授,博士生导师,研究方向:行为博弈、行为金融、行为公司治理.
  • 基金资助:

    教育部人文社科研究重大项目(02JAZD790010);教育部重点项目资助(06JHQZ0010)

A Compensation Ratio Game Equilibrium Study of China’s Non-Tradable Share Reform Based on EWA Learning Model

RAO Yu-lei, XU Yan-hui   

  1. Business School, Central South University, Changsha 410083, China
  • Received:2007-06-26 Revised:2007-12-23 Online:2008-02-28 Published:2008-02-28

摘要: 本文以股权分置改革中以纯送股方式支付对价的936家A股上市公司为样本,从行为博弈的角度,用EWA学习模型研究了其对价均衡的形成机理,并对均衡的形成过程进行了模拟。模拟结果表明EWA学习模型成功捕捉了对价均衡的形成过程,说明股权分置改革对价博弈是一个基于策略学习的博弈过程。对模型参数的分析表明:博弈双方即非流通股股东和流通股股东具有极强的学习能力,但这种学习仅为对过去经验的纯策略学习;同时非流通股股东制定方案时对策略收益赋予较小的权重,说明决策双方博弈地位的不对等在很大程度上决定了不公平对价均衡的形成。

关键词: 股权分置改革, 对价均衡, 行为博弈, EWA学习

Abstract: From the view of behavioral game theory,this paper uses EWA learning model to analyze the mechanism of the compensation ratio game,and simulates its equilibrium process. The simulation results from a sample of 936 reformed firms show that the EWA learning model successfully captured the equilibrium,which implicates that the process of the compensation game is based on strategy learning. By analyzing the model parameters,we find out that both non-tradable and tradable shareholders have strong learning ability in the decision-making,which is just an irrational learning of pure strategy from history experience,and the less weight placed on forgone payoffs by non-tradable shareholders shows that the unfair power in the game leads to the unfair equilibrium.

Key words: words ahereform of non-tradable share, the equilibrium of compensation ratio, behavioral game theory, EWA learning

中图分类号: