Research Into Some Problems On Nonlinear Characteristics And Forecast Theory Of Capital Market

This dissertation takes a close and deep look at nonlinear characteristics andforecast theory of Capital Market and its main subjects are as follows:1. Dissertating classical Capital Market theory and demonstrate that the classicalCapital market theory is of pattern of linearity. It took important role to accelerate thetheory of Finance, but it reveals a good many of limitation in describing exactly theperplexing behavior of Capital market today with the development of science andtechnology, especially with the development of nonlinear theory and technology.2. Summarizing the function and development of nonlinear science in CapitalMarket, such as the fractal theory, the chaos theory, wavelet theory and catastrophetheory.3. Having introduced the basic knowledge of fractal and the fractal market theory,researched into the difference between Fractal market theory and Effective markettheory, and pointed out that the Fractal Market theory can describe the Capital marketbehavior and researched into the essence of complexity on development andevolvement of Capital market very well.4. Investigating the test methods of the Fractal characteristics of Capital Market.The Empirical tests are conducted on the Fractal characteristics of the Chinese stockmarket. The results show that Shanghai and Shenzhen stock market have all obviousfractal characteristics, such as, self-similarity, state persistence, long-memory cycleand so on, which are not belong to that EMH described. Compared Shanghai withShenzhen market, whether date data or week data, the Hurst index of Shanghai marketis bigger than that of Shenzhen market. The departure of price variability of Shanghaimarket with Random Walk price is bigger than that of Shenzhen market. It indicatedthat the Fractal characteristic of Shanghai stock market is more obvious than Shenzhenstock market from the efficiency of security market.5. Introducing some basic concepts of Chaos theory, especial the basiccharacteristic of Chaos dynamic system, namely inherent randomness, butterfly effect,and strange attractor. The correlative dimension and the meaning of Lyapunovexponent are explained, the compute procedure is compiled, and the correlativedimension and the max Lyapunov exponent of Shanghai and Shenzhen Stock marketare computed empirically.6. Studied on the chaos behavior of Capital market and the determinability of chaoscharacteristic. The chaos characteristics of Chinese stock market are tested, theconclusion shows that the Shanghai and Shenzhen stock market have weak chaos.Studied nonlinearity characteristics on exchange rate of RMB to U.S. dollar, the result shows:exchange rate data of RMB to U S Dollar has all obvious fractal characteristics, but has notthe chaos phenomenon.7. Investigated the forecast methods of Capital market, especial the nonlinearforecast technology based on the theory of phase space reconstruction. The empiricaltest concerning the forecast effect of nonlinear auto-correlation chaos forecast modeland recursive nonlinear model is conducted, as example of Chinese stock market andexchange rate of Italy lira to U S Dollar. The results show that their forecast effect issatisfactory.