Papers › Graph Signal Processing for Global Stock Market Realized Volatility Forecasting
Graph Signal Processing for Global Stock Market Realized Volatility Forecasting
Zhengyang Chi, Junbin Gao, Chao Wang
This paper introduces an innovative realized volatility (RV) forecasting framework that extends the conventional Heterogeneous Auto-Regressive (HAR) model via integrating the Graph Signal Processing (GSP) technique. The volatility spillover effect is embedded and modeled in the proposed framework, which employs the graph Fourier transformation method to effectively analyze the global stock market dynamics in the spectral domain. In addition, convolution filters with learnable weights are applied to capture the historical mid-term and long-term volatility patterns. The empirical study is conducted with RV data of $24$ global stock market indices with around $3500$ common trading days from May 2002 to June 2022. The proposed model's short-term, middle-term and long-term RV forecasting performance is compared with various HAR type models and the graph neural network based HAR model. The results show that the proposed model consistently outperforms all other models considered in the study, demonstrating the effectiveness of integrating the GSP technique into the HAR model for RV forecasting.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
No leaderboard rows for this paper in the archive.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections