The information content of realized volatility of sector indices in China's stock market

This paper uses Ensemble Empirical Mode Decomposition (EEMD) to decompose and reconstruct the realized volatility into a high frequency component, a low frequency component and a trend. We find that, firstly, emerging industries have larger realized volatility than traditional industries. High frequency component contributes most to realized volatility, followed by low frequency component. Secondly, the trends of all sector indices first decreased and then increased from 2009 to 2016, but with the exception of information technology sector index. Thirdly, the low frequency components of A-share index, and emerging industries except financials sector index are related to coincident index and interest rate, with small fluctuations and long periods. While the low frequency components of financials sector index, and traditional industries except industrials sector index are related to consumer price index, exchange rate and M2, with large fluctuations and short periods. Finally, the long memory and leverage effect of emerging industries are shorter than those of traditional industries. The expected volumes of all sector indices have a monthly impact on the volatility. The daily and weekly unexpected volumes of traditional industries as well as the weekly unexpected volumes of emerging industries affect their realized volatility.