Research on long memory of realized volatility measurements in China Stock Market

By using the Detrended Fluctuation Analysis method, we study the long memory of the realized volatility and realized bi-power variation constructed from high-frequency intra-day return series of the Shenzhen stock index and its 14 component stocks in China Stock Market. We find that the realized volatility and realized bi-power variation exhibit significant long memory from one-minute to sixty-minute sampling interval. Moreover, with sampling interval increases, long memory of the realized volatility and realized bi-power variation weaken gradually.