Identifying market dynamics through the Hurst exponent

Fabrizio Di Sciorio; Laura Molero González; Juan E. Trinidad-Segovia; Fabrizio Di Sciorio; Laura Molero González; Juan E. Trinidad-Segovia

doi:10.3934/DSFE.2026004

Data Science in Finance and Economics

2026, Volume 6, Issue 1: 85-120. doi: 10.3934/DSFE.2026004

Previous Article Next Article

Research article

Identifying market dynamics through the Hurst exponent

Department of Economics and Business, University of Almería, 04120, Almería, Spain

Received: 19 March 2025 Revised: 18 December 2025 Accepted: 20 January 2026 Published: 27 February 2026
JEL Codes: C22, C45, C53, G17

In this study, we analyze a historical time series of crude oil prices to estimate the realized volatility and the Hurst exponent, with the objective of characterizing the temporal dynamics and persistence of the market. A clustering framework is employed to categorize historical realized volatility into three distinct regimes: low, moderate, and high volatility, thereby labeling observations accordingly. Subsequently, multiple machine learning models are implemented to predict these volatility regimes using the Hurst exponent as a key feature. The predictive performance of different algorithms is systematically evaluated to assess the effectiveness of the Hurst exponent as a volatility indicator. The findings suggest that the Hurst exponent provides a robust relative measure of market turbulence, demonstrating its potential as a predictive metric for financial market dynamics.
- market efficiency,
- clustering,
- turbulent market period,
- classification model,
- market dynamics,
- Hurst exponent
Citation: Fabrizio Di Sciorio, Laura Molero González, Juan E. Trinidad-Segovia. Identifying market dynamics through the Hurst exponent[J]. Data Science in Finance and Economics, 2026, 6(1): 85-120. doi: 10.3934/DSFE.2026004

Related Papers:

Abstract

In this study, we analyze a historical time series of crude oil prices to estimate the realized volatility and the Hurst exponent, with the objective of characterizing the temporal dynamics and persistence of the market. A clustering framework is employed to categorize historical realized volatility into three distinct regimes: low, moderate, and high volatility, thereby labeling observations accordingly. Subsequently, multiple machine learning models are implemented to predict these volatility regimes using the Hurst exponent as a key feature. The predictive performance of different algorithms is systematically evaluated to assess the effectiveness of the Hurst exponent as a volatility indicator. The findings suggest that the Hurst exponent provides a robust relative measure of market turbulence, demonstrating its potential as a predictive metric for financial market dynamics.

References

[1]	Ahn JJ, Oh KJ, Kim TY, et al. (2011) Usefulness of support vector machine to develop an early warning system for financial crisis. Expert Syst Appl 38: 2966–2973. https://doi.org/10.1016/j.eswa.2010.08.085 doi: 10.1016/j.eswa.2010.08.085
[2]	Akiba T, Sano S, Yanase T, et al. (2019) Optuna: A next-generation hyperparameter optimization framework. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2623–2631. https://doi.org/10.1145/3292500.333070
[3]	Alvarez-Ramirez J, Cisneros M, Ibarra-Valdez C, et al. (2002) Multifractal hurst analysis of crude oil prices. Physica A 313: 651–670. https://doi.org/10.1016/S0378-4371(02)00985-8 doi: 10.1016/S0378-4371(02)00985-8
[4]	Angelini D, Bianchi S (2023) Nonlinear biases in the roughness of a fractional stochastic regularity model. Chaos Soliton Fract 172: 113550. https://doi.org/10.1016/j.chaos.2023.113550 doi: 10.1016/j.chaos.2023.113550
[5]	Bašta M, Molnár P (2018) Oil market volatility and stock market volatility. Financ Res Lett 26: 204–214. https://doi.org/10.1016/j.frl.2018.02.001 doi: 10.1016/j.frl.2018.02.001
[6]	Bianchi S, Di Sciorio F, Mattera R (2022) Forecasting vix with hurst exponent In: Methods and Applications in Fluorescence. Cham: Springer International Publishing, Springer International Publishing, 90-95. https://doi.org/10.1007/978-3-030-99638-3_15
[7]	Bianchi S (2005) Pathwise identification of the memory function of multifractional brownian motion with application to finance. Int J Theor Appl Financ 8: 255–281. https://doi.org/10.1142/S0219024905002937 doi: 10.1142/S0219024905002937
[8]	Bianchi S, Pianese A (2014) Asset price modeling: From fractional to multifractional processes. Future perspect Risk Models Finance 18: 247–285. https://doi.org/10.1007/978-3-319-07524-2_7 doi: 10.1007/978-3-319-07524-2_7
[9]	Carmona P, Coutin L, Montseny G (2003) Stochastic integration with respect to fractional brownian motion. Ann Inst Henri Poincare-Probab Stat 39: 27–68. https://doi.org/10.1016/S0246-0203(02)01111-1 doi: 10.1016/S0246-0203(02)01111-1
[10]	Chatzis SP, Siakoulis V, Petropoulos A, et al. (2018) Forecasting stock market crisis events using deep and statistical machine learning techniques. Expert Syst Appl 112: 353–371. https://doi.org/10.1016/j.eswa.2018.06.032 doi: 10.1016/j.eswa.2018.06.032
[11]	Chen Y, Härdle WK, Pigorsch U (2010) Localized realized volatility modeling. J Am Stat Assoc 105: 1376–1393. https://doi.org/10.1198/jasa.2010.ap09039 doi: 10.1198/jasa.2010.ap09039
[12]	Cho P, Lee M (2022) Forecasting the volatility of the stock index with deep learning using asymmetric hurst exponents. Fractal Fract 6: 394. https://doi.org/10.3390/fractalfract6070394 doi: 10.3390/fractalfract6070394
[13]	Di Sciorio F (2023) Clustering analysis on hurst dynamic. Stud App Econ. https://doi.org/10.25115/sae.v42i1.9527
[14]	Di Sciorio F, Mattera R, Segovia JET (2023) Measuring conditional correlation between financial markets' inefficiency. Quant Financ Econ 7: 491–507. https://doi.org/10.3934/QFE.2023025 doi: 10.3934/QFE.2023025
[15]	Fernández-Martínez M, Sánchez-Granero MA, Muñoz Torrecillas MJ, et al. (2017) A comparison of three hurst exponent approaches to predict nascent bubbles in sp500 stocks. Fractals 25. https://doi.org/10.1142/S0218348X17500062
[16]	Granero MAS, Segovia JET, Pérez JG (2008) Some comments on hurst exponent and the long memory processes on capital markets. Physica A 387: 5543–5551. https://doi.org/10.1016/j.physa.2008.05.053 doi: 10.1016/j.physa.2008.05.053
[17]	Kristoufek L, Vosvrda M (2014a) Commodity futures and market efficiency. Energ Econ 42: 50–57. https://doi.org/10.1016/j.eneco.2013.12.001 doi: 10.1016/j.eneco.2013.12.001
[18]	Kristoufek L, Vosvrda M (2014b) Measuring capital market efficiency: long-term memory, fractal dimension and approximate entropy. Eur Phys J B 87: 1–9. https://doi.org/10.1140/epjb/e2014-50113-6 doi: 10.1140/epjb/e2014-50113-6
[19]	Kristoufek L, Vosvrda M (2016) Gold, currencies and market efficiency. Physica A 449: 27–34. https://doi.org/10.1016/j.physa.2015.12.075 doi: 10.1016/j.physa.2015.12.075
[20]	Lin WY, Hu YH, Tsai CF (2011) Machine learning in financial crisis prediction: a survey. IEEE Ieee T Syst Man Cy C 42: 421–436.
[21]	Mandelbrot BB (1982) The many faces of scaling: fractals, geometry of nature, and economics. In: Self-Organization and Dissipative Structures: Applications in the Physical and Social Sciences, 91–109. University of Texas Press.
[22]	Mandelbrot BB, Wallis JR (1969) Robustness of the rescaled range r/s in the measurement of noncyclic long run statistical dependence. Water Resour Res 5: 967–988. https://doi.org/10.1029/WR005i005p00967 doi: 10.1029/WR005i005p00967
[23]	Manokhin V (2022) Machine learning for probabilistic prediction. Phd thesis, Royal Holloway, University of London.
[24]	Mattera R, Di Sciorio F, Trinidad-Segovia JE (2022) A composite index for measuring stock market inefficiency. Complexity 1: 9838850. https://doi.org/10.1155/2022/9838850 doi: 10.1155/2022/9838850
[25]	Mensi W, Beljid M, Managi S (2014) Structural breaks and the time-varying levels of weak-form efficiency in crude oil markets: Evidence from the hurst exponent and shannon entropy methods. Int Econ 140: 89–106. https://doi.org/10.1016/j.inteco.2014.10.001 doi: 10.1016/j.inteco.2014.10.001
[26]	Peltier RF, Levy-Vehel J (1995) Multifractional brownian motion: Definition and preliminary results. Inria Res Rep.
[27]	Samitas A, Kampouris E, Kenourgios D (2020) Machine learning as an early warning system to predict financial crisis. Int Rev Financ Anal 71: 101507. https://doi.org/10.1016/j.irfa.2020.101507 doi: 10.1016/j.irfa.2020.101507
[28]	Yao CZ, Liu C, Ju WJ (2020) Multifractal analysis of the wti crude oil market, us stock market and epu. Physica A 550: 124096. https://doi.org/10.1016/j.physa.2019.124096 doi: 10.1016/j.physa.2019.124096

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)