HAR volatility modelling with jump

Heterogeneity and jump are two important factors influencing volatility. Ignoring heterogeneity and jump would reduce the quality of the realized volatility model. Underestimating downside risk is produced. In the present work we investigate the influence of heterogeneity and jump on volatility, establish a HAR-RV-J model based on the HAR-RV models. Our research demonstrates that the empirical results of HAR-RV-J model considering the jump effect presents better statistical indicators in the mainland stocks. Moreover, it is found that the forecasting performance of HAR-RV-J model is better than HAR-RV model by comparing RMSE, MAE and MAPE. It indicates that the combination of HAR-RV model and jump improves the estimation accuracy of the realized volatility in Chinese mainland stocks, which is more advantageous to the risk management of financial assets.