Research On Nonlinear Characteristics Of Gold Futures Market And Its Relevant Problems

Gold futures market is not only an extremely complex system but also a nonlinear system. However, the Efficient Market Hypothesis which describes the market basing on linearity hypothesis fails to accurately reflect the real market characteristics. On the contrary, the Fractal Market Hypothesis can be seem as the effective solution to the problem. This paper investigates the gold futures market of Shanghai futures exchange and New York commodity exchange, basing on the Fractal Market Hypothesis. By carrying out the statistical analysis and the empirical research of gold futures returns series, it quantitatively demonstrates that the two gold futures market actually has a fractal structure. In addition, in order to find out the similarities and differences in fractal characteristics of gold futures among the different markets and different times, the series are divided into several groups and studied separately so as to achieve more perfect research purposes.In the beginning, this paper tests the basic statistical characteristics and nonlinear characteristics of Shanghai and New York gold futures returns series. And based on the test results, further analysis of the fractal characteristics of gold futures returns was made, including single-fractal characteristics and multifractal characteristics analysis. In the single fractal characteristic analysis, this paper mainly use the Rescaled Range Analysis(R/S) to calculate the Hurst index of the gold futures returns series and fluctuation series to analyze the fractal and long memory characteristics of the gold futures market. Taking the existence of long memory characteristics in Shanghai and New York gold futures fluctuation series as premise, this paper carries out the measurement analysis of the gold futures by using FIGARCH model and FIAPARCH model with long memory characteristics and then finds out the most appropriate fluctuation model under the comparison of the models in normal distribution,t distribution, skewed-t distribution. Furthermore, in order to make up for the deficiency of single fractal analysis, further analysis of the multifractal characteristics of the gold futures returns is made by using Multifractal Detrended Fluctuation Analysis (MF-DFA). Along with the order q changing,, the Generalized Hurst index can show more detailed characterization of gold futures market such as durative change.Through the study of this paper, it shows that the use of Fractal Market Hypothesis and associated fractal analysis method for gold futures research is effective and it allows us to better understand the gold futures market intrinsic characteristics and evolution characteristic. All in all, this paper is a theoretical result which provides more theoretical basis for China's gold futures market's development.