Forecasting stock volatility during the stock market crash period: The role of Hawkes process

Abstract

We use a heterogeneous autoregressive model with Hawkes process (HAR-RV-H) to forecast the volatility of 300 major individual stocks in the Chinese stock market during the 2015 market crash period. The Hawkes intensity process is calculated with the tick-by-tick data of individual stocks. We show that the Hawkes indicator has predictive power for most individual stocks in the market crash period. We compare the in- and out-of-sample forecast results for the HAR type models and conclude that the Hawkes indicator can improve both in- and out-of-sample forecasting abilities.