Macro Factor Analysis Of Stock Price Fluctuation Based On EEMD Method

In the late 1990s,in order to get the better development of economy,China began with the attempt of stock market,now the development of the stock market in our country has obtained a good result,and its effect in the development of our country's economic has become more obvious and can not be ignored,it can say that the development of the stock market to a large extent affect the economic development in our country.And the development of the stock market will also directly affect the interests of investors,for example,the excessive volatility of stock prices will bring serious blow to the confidence of the investors to invest in stock market,affect the investors' point of view in the stock market and its investment,which affects the flow of the stock market's funds,affects the allocation of resources in securities market,and then do harm to the real economy's stability and development.In the research on China's stock market we found gradually that our country's stock market presents the obvious uncertainty,nonlinearity and instability,it is very difficult to use the traditional linear system analysis method to get satisfactory results in the study of stock price fluctuations.Norden E Huang(1998)proposed a new signal analysis method,the empirical mode decomposition(EMD),this method's appearance immediately caused wide public concern,and had been applied to a variety of academic fields.The EMD method is essentially based on the data sequence's characteristic time scales to get the sequence's decomposition,to obtain the intrinsic mode function of original data sequence,through the analysis of the intrinsic mode function to further analyse the original sequence's model of intrinsic fluctuations.This method does not need to set any basis function in advance,compared with the Fourier decomposition method,the wavelet decomposition approach,this method is going to reveal the original sequence from the essence of physical characteristics,and reflecting the characteristics of the data by this method is more real and accurate,and can more effectively keep all physical characteristics,has stronger local expression.However,the EMD method often has the problem of modal aliasing,so an improved algorithm is needed to overcome this problem and get more practical data decomposition results.This paper uses a collection of empirical mode decomposition(EEMD)method,which is an improved algorithm proposed for modal aliasing problem,the improved way is to add white noise in the original sequence,get a new sequence,then use EMD method to make decomposition of the new sequence.Using EEMD method to decompose the Shanghai composite index,this paper will decompose IMF components in preliminary statistical analysis,and reconstruct the IMF components based on a particular synthesis algorithm,then get a new low-frequency modal,a new high frequency modal,and a residual,and then analyze the economic meaning of the three components.Combining the EEMD method with the cointegration test theory in traditional measurement methods,the fluctuation characteristics and influencing factors of different frequency domain components of Shanghai composite index are analyzed and summarized.The analysis shows that the Shanghai composite index always fluctuates near the trend term,so it can be said that the trend term is the main component of stock price,reflecting the long-term trend of the stock market.The low-frequency mode reflects the long-term fluctuation of the stock market after the elimination of the trend.Macroeconomic development level and its impact on the stock market has the following characteristics:the impact range is wide,the degree of interference caused by macroeconomic is deeper,the mechanism that function in stock market is more complicated and causing larger stock price volatility.On the other hand,the characteristics and the height of the wave fit in the characteristics of low frequency mode correctly,so this paper considers that the low-frequency mode to a certain extent reflects the macroeconomic effect for long-term stock market fluctuations.The high frequency modality shows continuous small fluctuations,which partly reflects the impact of the short-term market's supply and demand imbalance and irregular events on the stock market price.Analyzing the low-frequency component(IMFL)of the Shanghai composite index,which on behalf of the macroeconomic influence factors of stock market fluctuations,and doing cointegration test of the IMFL sequence and the consumer price index(CPI),the measure of money supply growth(M2),industrial added value growth(GI),exchange rate(ER),interest rate(IR),analyzing the long-term equilibrium relationship between the six variables.The analysis of cointegration test results show that there is a long-term equilibrium relationship between the stock price and macroeconomic factors,the results also show that economic growth and money supply have positive impact on stock price fluctuations,however,inflation,exchange rate and interest rate have negative impact on stock price fluctuations.