Financial Date Processing Based On Hilbert-Huang Transformation

Since the open of Shanghai Stock Exchange in December 1990 and Shenzhen Stock Exchange in April 1991, China's stock market has been developing for about 20 years. The stock market is making rapid progress in the long run, though there has been bear market and bull market. Compared with the foreign market, facing the problem that the stock market lasts only a short period and the laws of it is not sound yet, both the government supervisor and the investers find it hard to know how the market is fluctuating. To do research on China's growing stock market and to understand the rule of it are of great theoretical and practical meanings.In this paper, the author utilizes the Hilbert-Huang Transformation (HHT) theory to process the Shanghai Stock Index(SSI), and the steps include applying Empirical Mode Decomposition(EMD) to the SSI, obtaining the period and the economic meaning of intrinsic mode function(IMF). At last, a method for predicting the SSI is introduced. The summary of the paper is as follows.1. Applying EMD to the SSI, drawing the figures of Hilbert Spectrum and Hilbert marginal spectrum, analysing the IMFs and the two spectrums. Computing the periods of the SSI and IMFs and making comparision of the three period-computed methods.2. Analysing kinds of economic variables associated to the SSI, finding those interrelated to the SSI, comparing the IMFs with the variables and finding the relation between them about period and fluctuation, illustrating the economic meanings of the IMFs.3. Utilising the Radial Basis Function (RBF) Artificial Neural Network(ANN) to predict the SSI according to the modified SSI that get rid of the high-frequenced IMFs and giving an error analysis.