Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets
We propose a dynamic fractional volatility model that incorporates a time-varying Hurst exponent estimated via Daubechies-4 wavelet analysis on 252-day rolling windows to capture evolving market memory effects in equity markets. This approach overcomes the limitations of traditional GARCH-type and s...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-06-01
|
| Series: | Frontiers in Applied Mathematics and Statistics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fams.2025.1554144/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839635996092661760 |
|---|---|
| author | Abe Webb Siddharth Mahajan Mateo Sandhu Rohan Agarwal Arjun Velan |
| author_facet | Abe Webb Siddharth Mahajan Mateo Sandhu Rohan Agarwal Arjun Velan |
| author_sort | Abe Webb |
| collection | DOAJ |
| description | We propose a dynamic fractional volatility model that incorporates a time-varying Hurst exponent estimated via Daubechies-4 wavelet analysis on 252-day rolling windows to capture evolving market memory effects in equity markets. This approach overcomes the limitations of traditional GARCH-type and static fractional volatility models, which assume a constant memory parameter and struggle during regime shifts and market stress. The model is applied to daily closing prices of the S&P 500 Index over 1,258 trading days from January 1, 2015 to December 31, 2019, yielding statistically significant improvements in forecasting performance. Empirical results indicate a 12.3 % reduction in RMSE and a 9.8 % improvement in MAPE, with an out-of-sample R-squared exceeding 0.72 compared to benchmark models. Maximum likelihood estimation with Fisher scoring is used for daily parameter updates, ensuring the model remains responsive to rapidly changing market conditions. Additionally, the model achieves an average absolute option pricing error of 1.8 %, markedly lower than that of traditional specifications. These enhancements are further corroborated by pairwise Diebold–Mariano tests, which confirm the statistical significance of the forecast improvements. Overall, this framework offers a rigorous and computationally efficient method for real-time volatility forecasting that delivers substantial benefits for risk management, derivative pricing, and automated trading strategies, grounded in robust statistical methodology. |
| format | Article |
| id | doaj-art-1e55d2e77c8844d9ab690660ac7f6507 |
| institution | Matheson Library |
| issn | 2297-4687 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Applied Mathematics and Statistics |
| spelling | doaj-art-1e55d2e77c8844d9ab690660ac7f65072025-07-08T10:11:21ZengFrontiers Media S.A.Frontiers in Applied Mathematics and Statistics2297-46872025-06-011110.3389/fams.2025.15541441554144Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity marketsAbe WebbSiddharth MahajanMateo SandhuRohan AgarwalArjun VelanWe propose a dynamic fractional volatility model that incorporates a time-varying Hurst exponent estimated via Daubechies-4 wavelet analysis on 252-day rolling windows to capture evolving market memory effects in equity markets. This approach overcomes the limitations of traditional GARCH-type and static fractional volatility models, which assume a constant memory parameter and struggle during regime shifts and market stress. The model is applied to daily closing prices of the S&P 500 Index over 1,258 trading days from January 1, 2015 to December 31, 2019, yielding statistically significant improvements in forecasting performance. Empirical results indicate a 12.3 % reduction in RMSE and a 9.8 % improvement in MAPE, with an out-of-sample R-squared exceeding 0.72 compared to benchmark models. Maximum likelihood estimation with Fisher scoring is used for daily parameter updates, ensuring the model remains responsive to rapidly changing market conditions. Additionally, the model achieves an average absolute option pricing error of 1.8 %, markedly lower than that of traditional specifications. These enhancements are further corroborated by pairwise Diebold–Mariano tests, which confirm the statistical significance of the forecast improvements. Overall, this framework offers a rigorous and computationally efficient method for real-time volatility forecasting that delivers substantial benefits for risk management, derivative pricing, and automated trading strategies, grounded in robust statistical methodology.https://www.frontiersin.org/articles/10.3389/fams.2025.1554144/fulltime-varying Hurst exponentvolatility modelingfractal dynamicswavelet analysisadaptive market hypothesisstochastic volatility |
| spellingShingle | Abe Webb Siddharth Mahajan Mateo Sandhu Rohan Agarwal Arjun Velan Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets Frontiers in Applied Mathematics and Statistics time-varying Hurst exponent volatility modeling fractal dynamics wavelet analysis adaptive market hypothesis stochastic volatility |
| title | Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets |
| title_full | Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets |
| title_fullStr | Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets |
| title_full_unstemmed | Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets |
| title_short | Adaptive fractal dynamics: a time-varying Hurst approach to volatility modeling in equity markets |
| title_sort | adaptive fractal dynamics a time varying hurst approach to volatility modeling in equity markets |
| topic | time-varying Hurst exponent volatility modeling fractal dynamics wavelet analysis adaptive market hypothesis stochastic volatility |
| url | https://www.frontiersin.org/articles/10.3389/fams.2025.1554144/full |
| work_keys_str_mv | AT abewebb adaptivefractaldynamicsatimevaryinghurstapproachtovolatilitymodelinginequitymarkets AT siddharthmahajan adaptivefractaldynamicsatimevaryinghurstapproachtovolatilitymodelinginequitymarkets AT mateosandhu adaptivefractaldynamicsatimevaryinghurstapproachtovolatilitymodelinginequitymarkets AT rohanagarwal adaptivefractaldynamicsatimevaryinghurstapproachtovolatilitymodelinginequitymarkets AT arjunvelan adaptivefractaldynamicsatimevaryinghurstapproachtovolatilitymodelinginequitymarkets |