🤖 AI Summary
To address insufficient predictive accuracy for stock market realized volatility, this paper proposes the HAR-PD model family, which integrates path-dependence characteristics into the Heterogeneous Autoregressive (HAR) framework—marking the first incorporation of path-dependent volatility decomposition into HAR modeling. The core innovation is the HAR-REQ model, which dynamically identifies trend and reversal patterns in price paths using empirically determined quantile thresholds, thereby explicitly capturing volatility’s long- and short-term memory effects and asymmetric responses. Empirical analysis on high-frequency data from the Shanghai and Shenzhen stock markets demonstrates that HAR-PD models significantly outperform the benchmark HAR model, reducing average MAE by 12.6% and RMSE by 11.3%. Robustness is confirmed through rolling-window estimation, subsample analysis, and alternative volatility measures. This work introduces a novel path-dependent perspective to volatility modeling and provides an interpretable, statistically grounded toolkit for practitioners and researchers.
📝 Abstract
Volatility forecasting in financial markets is a topic that has received more attention from scholars. In this paper, we propose a new volatility forecasting model that combines the heterogeneous autoregressive (HAR) model with a family of path-dependent volatility models (HAR-PD). The model utilizes the long- and short-term memory properties of price data to capture volatility features and trend features. By integrating the features of path-dependent volatility into the HAR model family framework, we develop a new set of volatility forecasting models. And, we propose a HAR-REQ model based on the empirical quartile as a threshold, which exhibits stronger forecasting ability compared to the HAR-REX model. Subsequently, the predictive performance of the HAR-PD model family is evaluated by statistical tests using data from the Chinese stock market and compared with the basic HAR model family. The empirical results show that the HAR-PD model family has higher forecasting accuracy compared to the underlying HAR model family. In addition, robustness tests confirm the significant predictive power of the HAR-PD model family.