Graph Signal Processing for Global Stock Market Realized Volatility Forecasting

Zhengyang Chi, Junbin Gao and Chao Wang

Papers from arXiv.org

Abstract: 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.

Date: 2024-10, Revised 2025-02
New Economics Papers: this item is included in nep-fmk and nep-mac
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://arxiv.org/pdf/2410.22706 Latest version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:arx:papers:2410.22706

Access Statistics for this paper

More papers in Papers from arXiv.org
Bibliographic data for series maintained by arXiv administrators ().