Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation

This study proposes a novel hybrid model for stock volatility forecasting by integrating directional and temporal dependencies among financial time series and market regime changes into a unified modeling framework. Specifically, we design a novel Hurst Exponent Effective Transfer Entropy Graph Neur...

Full description

Saved in:
Bibliographic Details
Main Authors: Sangheon Lee, Poongjin Cho
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Fractal and Fractional
Subjects:
Online Access:https://www.mdpi.com/2504-3110/9/6/339
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850167703215538176
author Sangheon Lee
Poongjin Cho
author_facet Sangheon Lee
Poongjin Cho
author_sort Sangheon Lee
collection DOAJ
description This study proposes a novel hybrid model for stock volatility forecasting by integrating directional and temporal dependencies among financial time series and market regime changes into a unified modeling framework. Specifically, we design a novel Hurst Exponent Effective Transfer Entropy Graph Neural Network (H-ETE-GNN) model that captures directional and asymmetric interactions based on Effective Transfer Entropy (ETE), and incorporates regime change detection using the Hurst exponent to reflect evolving global market conditions. To assess the effectiveness of the proposed approach, we compared the forecast performance of the hybrid GNN model with GNN models constructed using Transfer Entropy (TE), Granger causality, and Pearson correlation—each representing different measures of causality and correlation among time series. The empirical analysis was based on daily price data of 10 major country-level ETFs over a 19-year period (2006–2024), collected via Yahoo Finance. Additionally, we implemented recurrent neural network (RNN)-based models such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) under the same experimental conditions to evaluate their performance relative to the GNN-based models. The effect of incorporating regime changes was further examined by comparing the model performance with and without Hurst-exponent-based detection. The experimental results demonstrated that the hybrid GNN-based approach effectively captured the structure of information flow between time series, leading to substantial improvements in the forecast performance for one-day-ahead realized volatility. Furthermore, incorporating regime change detection via the Hurst exponent enhanced the model’s adaptability to structural shifts in the market. This study highlights the potential of H-ETE-GNN in jointly modeling interactions between time series and market regimes, offering a promising direction for more accurate and robust volatility forecasting in complex financial environments.
format Article
id doaj-art-6268ff0ced7a4265b5ca65e4af94288d
institution OA Journals
issn 2504-3110
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Fractal and Fractional
spelling doaj-art-6268ff0ced7a4265b5ca65e4af94288d2025-08-20T02:21:09ZengMDPI AGFractal and Fractional2504-31102025-05-019633910.3390/fractalfract9060339Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime AdaptationSangheon Lee0Poongjin Cho1School of Computing, Gachon University, Seongnam 13120, Republic of KoreaSchool of Computing, Gachon University, Seongnam 13120, Republic of KoreaThis study proposes a novel hybrid model for stock volatility forecasting by integrating directional and temporal dependencies among financial time series and market regime changes into a unified modeling framework. Specifically, we design a novel Hurst Exponent Effective Transfer Entropy Graph Neural Network (H-ETE-GNN) model that captures directional and asymmetric interactions based on Effective Transfer Entropy (ETE), and incorporates regime change detection using the Hurst exponent to reflect evolving global market conditions. To assess the effectiveness of the proposed approach, we compared the forecast performance of the hybrid GNN model with GNN models constructed using Transfer Entropy (TE), Granger causality, and Pearson correlation—each representing different measures of causality and correlation among time series. The empirical analysis was based on daily price data of 10 major country-level ETFs over a 19-year period (2006–2024), collected via Yahoo Finance. Additionally, we implemented recurrent neural network (RNN)-based models such as Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) under the same experimental conditions to evaluate their performance relative to the GNN-based models. The effect of incorporating regime changes was further examined by comparing the model performance with and without Hurst-exponent-based detection. The experimental results demonstrated that the hybrid GNN-based approach effectively captured the structure of information flow between time series, leading to substantial improvements in the forecast performance for one-day-ahead realized volatility. Furthermore, incorporating regime change detection via the Hurst exponent enhanced the model’s adaptability to structural shifts in the market. This study highlights the potential of H-ETE-GNN in jointly modeling interactions between time series and market regimes, offering a promising direction for more accurate and robust volatility forecasting in complex financial environments.https://www.mdpi.com/2504-3110/9/6/339forecastingvolatilityHurst exponenteffective transfer entropygraph neural network
spellingShingle Sangheon Lee
Poongjin Cho
Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
Fractal and Fractional
forecasting
volatility
Hurst exponent
effective transfer entropy
graph neural network
title Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
title_full Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
title_fullStr Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
title_full_unstemmed Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
title_short Graph-Based Stock Volatility Forecasting with Effective Transfer Entropy and Hurst-Based Regime Adaptation
title_sort graph based stock volatility forecasting with effective transfer entropy and hurst based regime adaptation
topic forecasting
volatility
Hurst exponent
effective transfer entropy
graph neural network
url https://www.mdpi.com/2504-3110/9/6/339
work_keys_str_mv AT sangheonlee graphbasedstockvolatilityforecastingwitheffectivetransferentropyandhurstbasedregimeadaptation
AT poongjincho graphbasedstockvolatilityforecastingwitheffectivetransferentropyandhurstbasedregimeadaptation