摘要
刻画了混沌经济系统信息耗散尺度.计算深圳股指日、周、月对数收益率时序的关联维数、二阶Renyi熵及最佳嵌入维数.结果表明文中的参数选取及算法应用是好的.比较了不同时序K2熵的性态,研究了信息耗散尺度指标演化规律及其形成机制,解释了存在与经济系统信息耗散尺度演变背后的经济学原因.研究结果表明对正相关分形时序(FBM)在较长的时间间隔内用较大的尺标观测将导致信息加速丧失,而负相关FBM正相反.
Scale of information dissipation in chaos economic system is portrayed. The correlative dimension, Renyi entropy (K2) and optimized embedding dimension is calculated, which can be used for analyzing chaos time-series of logarithmic yield of day, week, month in the Shenzhen stock market. The result indicates that the arithmetic application and the select of parameter is efficient. The character of K2 the different time-series is analyzed. The evolvement rule, mechanism and economic causation is researched and made clear. The result indicates that the information is accelerated losed to positive correlative fractal time-series when it is observed with the bigger scale in the longer interval, and it is reverse to negative fractal time-series.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
2005年第5期20-26,共7页
Systems Engineering-Theory & Practice
基金
信息管理与信息经济学教育部重点实验室开放基金资助(F04 19)
关键词
证券市场
信息耗散尺度
混沌
关联维数
RENYI熵
机制
securities market
scale of information dissipation
chaos
correlative dimension
Renyi entropy
mechanism
