The complexity and fractal structures of CSI300 before and after the introduction of CSI300IF

Abstract

This paper employs multifractal detrended fluctuation analysis (MF-DFA) approach to analyze the complexity and fractal structures of China Securities Index 300 (CSI300), which covers the period from April 5, 2006 to May 9, 2014. Through comparing the statistic properties before and after the introduction of CSI300 index futures (CSI300IF), we find that: (1) the price return series exhibits multifractal properties for the two periods; (2) by comparing the efficient measure based on the general Hurst exponent, we find the market becomes more efficient after the introduction of CSI300IF; (3) the width of multifractal spectrum becomes narrower, which means the complexity of the market is decreased. Therefore, the introduction of CSI300IF is beneficial to reduce the risk and increase the efficiency of Chinese securities market. Furthermore, the main sources of multifractality of these time series are examined through the shuffle method, and the results show that before CSI300IF the multifractality is mainly due to the long-range correlation properties between small and large fluctuations, while after CSI300IF it is more due to fat-tailed probability distributions.