Akademik Veri Yönetim Sistemi
First International Meeting on Behavioral Social Sciences, Almería, İspanya, 23 - 24 Nisan 2026, ss.37, (Özet Bildiri)
This study demonstrates effectiveness of Generalised Hurst Exponent (GHE) based statistical
arbitrage strategy on high-volatile cryptocurrency market, using high-frequency data for the
2022–2023 period. The main objective is to evaluate the ability of the GHE approach to identify
profitable investment opportunities in an environment characterised by high volatility [1].
The dataset involves minute-by-minute closing prices of the 20 cryptocur- rencies with the
highest market capitalisation, traded against Bitcoin on the Binance exchange. These
cryptocurrencies represent approximately 90% of the total market capitalisation, which
guarantees the representativeness of the results [2]. The methodology is structured in three
phases: price normalisation and construction of the pair spread by minimising the Hurst exponent
(H) [3], weekly selection of pairs with H < 0.5, indicative of anti-persistent behaviour and
potential reversion to the mean [4], and execution of a trading strategy based on entry and exit
rules defined by standard deviations from the moving average of the spread [5]. We will include
transaction costs of 0.01% to avoid bias in overestimating the returns generated. The empirical
results show that the GHE-based strategy achieves pos- itive and stable annualised returns for
most of the period analysed. Like- wise, risk-adjusted performance indicators, such as the Sharpe
ratio and the Sortino ratio, show high values, while the maximum drawdown remains at low
levels, demonstrating adequate risk management. The coefficient of de- termination (R2) remains
close to unity, indicating a high explanatory power of the model. In conclusion, the study
validates the viability and robustness of the GHE as a pair selection tool in cryptocurrency
markets, highlighting its ability to adapt to high-frequency conditions. The findings provide
relevant empirical evidence for the quantitative finance literature and offer.
Keywords: Statistical Arbitrage, Cryptocurrencies, High-frequency, Generalized Hurst Exponent.
References
Ramos-Requena, J. P., García Amate, A., & López-García, M. D. L. N. (2025). Statistical
Arbitrage: An Approach from Econophysics. In Ad- vances in Quantitative Methods for
Economics and Business: A Tribute to José García Pérez (pp. 313-333). Cham: Springer Nature
Switzerland.
Bağcı, M., & Kaya Soylu, P. (2025). The optimal threshold selection for high-frequency pairs
trading via supervised machine learning algo- rithms. Computational Economics, 1-29
Barunik, J., & Kristoufek, L. (2010). On Hurst exponent estimation under heavy-tailed
distributions. Physica A: statistical mechanics and its applications, 389(18), 3844-3855.
Ramos-Requena, J. P., Trinidad-Segovia, J. E., & Sánchez-Granero, M. A. (2017). Introducing
Hurst exponent in pair trading. Physica A: statistical mechanics and its applications, 488, 39-45.
Ramos-Requena, J. P., Trinidad-Segovia, J. E., & Sánchez-Granero, M. Á. (2020). Some notes
on the formation of a pair in pairs trading. Mathematics, 8(3), 348.