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 ›› 2011, Vol. 19 ›› Issue (4): 9-16.

• 论文 • 上一篇    下一篇

部分信息下实物期权的定价和风险对冲

杨金强1,2, 杨招军1   

  1. 1. 湖南大学金融与统计学院, 湖南 长沙 410079;
    2. 上海财经大学金融学院, 上海 200433
  • 收稿日期:2010-01-27 修回日期:2011-06-21 出版日期:2011-08-30 发布日期:2011-08-30
  • 作者简介:杨金强(1983- ),男(汉族),河北衡水人,湖南大学金融与统计学院博士研究生,上海财经大学金融学院助理教授,研究方向:数理金融、金融工程.
  • 基金资助:

    国家自然科学基金项目(70971037);教育部博士点基金课题(20100161110022);湖南省研究生科研创新项目(CX2009B064)

The Pricing of Real Option and Risk Hedging under Partial Information

YANG Jin-qiang1,2, YANG Zhao-jun1   

  1. 1. School of Finance and Statistics, Hunan Unitersity, Changsha 410079, China;
    2. School of Finance, Shanghai Unitersity of Finance and Economics, Shanghai 200433, China
  • Received:2010-01-27 Revised:2011-06-21 Online:2011-08-30 Published:2011-08-30

摘要: 当前所有实物期权理论研究都是基于完全信息(full information)假设.本文则通过研究投资者在部分信息(partial information)下极大化无限期消费效用的最优投资消费问题,得出实物期权的消费效用无差别价格.通过控制系统的分离原理,运用Kalman滤波技术和随机控制方法,得到了CARA效用函数情形下实物期权的自由边界偏微分方程.利用有限差分法,解得实物期权的隐含价值及最优执行水平从而得到最优投资消费策略和效用函数的数值解.通过蒙特卡洛模拟,给出了投资者在完全信息和部分信息下的动态决策差异,并且通过比较两种信息水平下的投资者福利给出了信息价值的测算.

关键词: 实物期权, 部分信息, Kalman滤波, 信息价值

Abstract: The current real option pricing theory is based on the full information.In this paper,we relax this assumption and consider the optimal control problem of investment and consumption during an infinite horizon to explore the consumption-utility based indifference price of real option with partial observation,which is known as partial information.Thanks to Kalman filtering,dynamic programming and Hamilton-Jacobi-Bellman theory,an implied option value is given by the semi-closed-form solution to the two free-boundary PDE under the separation principle of control system and the numerical results are obtained by the finite difference method.Further more,by Monte Carlo simulation,the difference of strat egies between partial information and full information is discussed.Finally,we explore the relation of the two value functions under partial and full observation,and the economic value of information is derived.

Key words: real option, partial information, Kalman filtering, value of information

中图分类号: