Science, Technologies, Innovations №1(37) 2026, 68-79 р

https://doi.org/10.35668/2520-6524-2026-1-07

Shabelnyk T. V. — D. Sc. in Economics, Professor, Chair of the Department of Economic Cybernetics and Systems Analysis Simon Kuznets Kharkiv National University of Economics, 9A, Nauky Ave, Kharkiv, Ukraine, 61001; +38 (050) 176-84-78; Tanya.shabelnik17@gmail.com; ORCID: 0000-0001-9798-391X

Yevsyeyeva S. O. — Student, Simon Kuznets Kharkiv National University of Economics, 9A, Nauky Ave., Kharkiv, Ukraine, 61001; +38 (050) 176-84-78; Sofiia.Yevsieieva@hneu.net; ORCID: 0009-0001-9496-6804

QUANTITATIVE MODELING OF PORTFOLIO STRATEGIES USING MACHINE LEARNING-BASED ALPHA SIGNALS AND BACKTESTING

Abstract. This paper proposes a quantitative framework for constructing ML-based alpha signals and integrating them into a portfolio backtesting procedure over a medium-term investment horizon. The empirical analysis relies on daily financial time series of liquid equity market instruments, providing statistically consistent input data for quantitative modeling. A structured feature space is constructed, incorporating price-based, volatility-based, and intraday-aggregated characteristics. Model training is implemented in a time-aware setting using rolling or expanding window schemes and strict chronological separation of the training, validation, and test sets, ensuring robust quantitative out-of-sample evaluation and minimizing information leakage.
Alpha signals are generated at discrete rebalancing dates as numerical forecasts of expected returns over a predefined horizon and are directly embedded into portfolio construction. Performance is assessed through quantitative portfolio backtesting using standard risk-adjusted metrics, including returns, volatility, the Sharpe ratio, and maximum drawdown. The proposed algorithmic pipeline formalizes the interaction between machine learning and portfolio analysis and provides a methodological basis for the quantitative evaluation of the investment performance of ML-based strategies.

Keywords: quantitative modeling, machine learning, alpha signals, portfolio backtesting, equity market, mediumterm strategies, risk-adjusted performance metrics.

REFERENCES

  1. Gu, S., Kelly, B., & Xiu, D. (2020). Empirical Asset Pricing via Machine Learning, The Review of Financial Studies, 33 (5), 2223-2273. DOI: https://doi.org/10.1093/rfs/hhaa009.
  2. Bagnara, (2024). Asset Pricing and Machine Learning: A critical review. Journal of Economic Sur-Veys, 38, 27-56. DOI: 10.1111/joes.12532.
  3. Lopez de Prado, (2018). Advances in Financial Machine Learning. Hoboken, NJ: John Wiley & Sons, 400 p.
  4. Harvey, C. R., & Liu, Y. (2015). B Available at SSRN 2345489.
  5. Bailey, H., & Lopez de Prado, M. (2012). The Sharpe ratio efficient frontier. The Journal of Risk, 15 (2), 3-44. DOI: https://doi.org/10.21314/JOR.2012.255.
  6. Nadler, , & Sancetta, A. (2023). Empirical Asset Pricing with Functional Factors. Journal of Financial Econometrics, 21 (5), 1258-1281. DOI: https://doi.org/10.1093/jjfinec/nbac003.
  7. Wang, (2025). Machine learning for stock return prediction: Transformers or simple neural networks? Finance Research Letters, 86 (Part F), 108783. DOI: https://doi.org/10.1016/j.frl.2025.108783.
  8. Giglio, S., & Xiu, D. (2021). Asset pricing with omitted factors. Journal of Political Economy, 129 (7), 1947-1990. Retrieved from: https://www.journals.uchicago.edu/doi/10.1086/714090.
  9. Potrashkova, L., Zaruba, V., Raiko, , & Yevsyeyev, O. (2024). Identifying the system of value factors of green consumer choice. Innovative Marketing, 20 (1), 199-211. DOI: http://dx.doi.org/10.21511/im.20(1).2024.17.
  10. Maslyhan, O., Shabelnyk, T., Korolovych, , & Liba, N. (2022), Modern Approach to Modeling Of Efficiency Of Financial Market Based On Methods Of Dynamic Programming. Elektronnyi zhurnal «Efektyvna ekonomika» [Electronic magazine “Effective Economy“], 9. DOI: http://doi.org/10.32702/2307-2105.2022.9.7.
  11. Bulatova, , Shabelnyk, T., Marena, T., & Reznikova, N. (2019). Influence of regional financial market models on the structure of global financial assets. Advances in Economics, Business and Management Research (AEBMR), 95. 6th International Conference on Strategies, Models and Technologies of Economic Systems Management (SMTESM 2019). Proceedings of the International Scientific Conference, P. 339-345. DOI: https://doi.org/10.2991/smtesm-19.2019.55.
  12. Shabelnyk, V. (2018). Approaches to optimize investment risks. In Proceedings of the International Conference “Problems of Decision Making Under Uncertainties” (Con-ference Materials), 2018.
  13. Wang, Y., Huang,, & Luo, J. (2025). Predicting Stock Prices Based on Machine Learning to Build Self-adaptive Trading Strategy. Computational Economics. DOI: 10.1007/s10614-025-11054-4.
  14. Akyildirim, , Nguyen, D. K., Sensoy, A., & Šikić, M.(2023). Forecasting high-frequency excess stock returns via data analytics and machine learning. International Review of Financial Analysis, 29 (1), 22-75. DOI: https://doi.org/10.1111/eufm.12345.