Forecasting international equity market volatility: A new approach

We propose a new heterogeneous autoregressive (HAR) model to investigate whether the novel HAR‐RV model we design exhibits superior predictive ability compared with the classical HAR‐RV model in forecasting the realized volatility (RV) of international equity markets. We reconstruct the weekly and monthly HAR components by using the exponentially weighted moving averages (EWMA) method, and the new model is therefore termed EWMA‐HAR‐RV. The in‐sample results and the out‐of‐sample analyses suggest that the EWMA‐HAR‐RV model exhibits stronger predictive power in predicting most international equity market RVs. We further analyze the source of additional predictive power for the EWMA‐HAR‐RV model and find that giving more weight to recent information and using more historical information are helpful for the predictions.