Forecasting realized volatility based on the truncated two-scales realized volatility estimator (TTSRV): Evidence from China's stock market

In this paper, the TTSRV (truncated two-scale realized volatility estimator), a novel estimator of the continuous part of realized volatility (RV), is used to forecast the RV of the SSEC index. Based on the classic heterogeneous autoregressive model for RV (HAR-RV), our new model, which applies the TTSRV, can describe the continuous and jump processes of RV with higher accuracy. The empirical results obtained by this study suggest that the TTSRV outperforms previous models in both statistical and economic aspects.