Chaos time series analysis has become a hot topic in the research on economy and finance systems. And the nonlinear chaos characteristic research of economy and finance time series is the key of the research,which is vital to the research of phase space reconstructing and modeling in the analysis of the economy and finance. It affects the efficiency of modeling and the effect of forecasting directly, so the research of non-linear chaos characteristic is more and more important.This thesis introduces the complex science and its background in detail firstly, and put forward a detailed summary about the present progress in the domestic and foreign similar research. And then, in the foundation of the overseas scholars'study, I study RP and the CRP methods for visualization the dynamics characteristic of the phase space track as well as the complexity degree algorithm of chaos time series——ApEn algorithm. Applying these two methods based on the VB program we analyze separately the data which is obtained from complex financial system model and Lorenz system, and obtain several new recurrence plots and cross recurrence plots. This provides the graph gist of determination complex intrinsic characteristic. The results establish the foundation for exact and further division of chaos time series.And then, the thesis still carries out in-depth analysis for the related dimension of chaos time series, and makes some beneficial exploration about the relations between related dimension, embedded- dimension and critical distance. In order to study the nonlinear characteristic of chaos time series, we use the data of the RMB to the American dollar exchange rate to study embedded-dimension, time delay, reconstructing the phase space, reducing noise, fractal, Lyapunov exponent, BDS statistic and so on to decide its important nonlinear characteristic.Then we gain some beneficial results applying the correlative theories and methods through Canadian to American dollar exchange rate data.The research provides profitable reference to the serial research of economy and finance data, and the research achievement has important academic and practical meanings. |