Testing Nonlinear Structure and Chaos Effect of Chinese Stock Markets Based on BDS and CR Methods

马超群,邹琳,李红权
2008-01-01
Abstract:Based on the analysis of the traditional methods,the small data algorithm originated by Rosenstein was introduced to calculate the maximum Lyapunov exponent,and two types of tests-BDS and close return(CR) were used to further analyze the dynamical characteristic of chaotic of Chinese stock markets.For avoiding destroying the fractal structure,we utilized a special method of log-linear detrended(LLD) to process the sample data.The conclusions indicate that the notable chaos dynamics characteristic appear to exhibit in Chinese stock markets,such as a low dimensionality chaos attractor,sensitive dependence on initial values,quasi-periodicity.And the economical implication and application value of chaos was also investigated.
What problem does this paper attempt to address?