Microscopic characteristics of Chinese capital market based on the high frequency data of Shanghai composite index
Tang Zhen-Peng,Chen Wei-Hong,Ran Meng,,,
DOI: https://doi.org/10.7498/aps.66.120203
IF: 0.906
2017-01-01
Acta Physica Sinica
Abstract:This paper mainly uncovers the typical microscopic characteristics of Chinese capital market in three different stock price stages of rising, steady and falling based on the high frequency data of Shanghai composite index. Firstly, by analyzing the probability distribution of the Shanghai composite index in different time intervals, we clearly find that the logarithmic change of the index presents an obvious heavy tail feature as well as non-Gaussian Levy distribution, and the return series converges to a normal distribution with the increase of the time interval, which becomes more significant especially in the falling stage of stock prices. Secondly, by calculating the autocorrelation function, we observe that unlike the return rate, the fluctuation ratio of Shanghai composite index demonstrates remarkable long memory volatility with a periodicity of about 240 min, and the autocorrelation curve in falling stage is much higher than in rising and steady stages. Thirdly, in the multi-fractal structure, the volatility of return series has significant short-term and long-term differences among three different stages of rising, steady and falling due to the effects of time limitation and liquidity of investment. Finally, the macroscopic behavior of the Shanghai composite index is relatively consistent with that of the international mature stock market, however, the corresponding microscopic characteristics demonstrate significant differences due to the fact that the Chinese capital market is strongly dependent on the macroeconomic policy, investor sentiment, and liquidity levels. It is quite remarkable that the tail distribution of mature stock market is much fatter than that of Chinese stock market because of the special control and limit mechanism of stock prices in China, which finally causes the considerably lower amplitude of price fluctuation. Moreover, it is also found that the attenuation speed of the autocorrelation function in the Chinese capital market is obviously slower than that in the mature stock market, which suggests that the behaviors of investors in Chinese stock market are more likely to be influenced by the historic exchange information. At the same time, the periodicity of autocorrelation function is actually caused by the inertia recoil of investors, which further verifies the information asymmetry of Chinese stock market. Especially, by changing the starting values of the samples, we find that the periodicity of autocorrelation function still remains the same, which indicates that the periodicity characteristic of stock price is not dominated only by the intraday pattern of trading activity. Therefore, the investors should discover the underlying rules of high-frequency data and extract more useful knowledge in order to guide their investment decisions more effectively.
physics, multidisciplinary