The Impact of Artificial Intelligence on Portfolio Performance: A Comparative Study of Machine Learning and Traditional Portfolio Optimisation Approaches

Author

Shaik Reshma , Mr. V. Narendra


Abstract

 Artificial intelligence and machine learning are now being used in investment products, but it remains uncertain whether these tools consistently outperform established portfolio optimisation methods or control risk more effectively. This study compares three AI/ML-oriented exchange-traded funds—AIEQ, QRFT and AMOM—with three funds following minimum-variance or low-volatility strategies—USMV, SPLV and ACWV. The comparison covers 2021–2025 and relies on secondary data for annual total returns, the S&P 500 total-return benchmark and the three-month Treasury bill rate. Performance is assessed using returns, Sharpe and Sortino ratios, and the worst annual loss. Risk-management efficiency is represented by a combined measure based on volatility, downside capture, worst-year loss and beta. The analysis uses chi-square tests, one-way and two-way ANOVA, multiple regression and mediation testing. Traditional funds recorded stronger risk-management efficiency (F = 67.9, p = 0.001), while the difference in average annual returns between the groups was not statistically significant (10.8% for ML-based funds and 7.8% for traditional funds; p = 0.622). Although the regression model was significant overall (R² = 0.76), market conditions were the only significant individual predictor. Results also indicate that the relative performance of the two approaches changes with the market: ML-based funds did better in rising periods but experienced substantially larger losses in 2022. Risk-management efficiency did not significantly explain returns, and the proposed mediation effect was not supported. Given the limited sample, these findings should be treated as preliminary rather than conclusive.



Keywords

artificial intelligence, machine learning, portfolio optimisation, minimum variance, risk management efficiency, Sharpe ratio, portfolio performance.



Full Text:

Download Paper PDF


References


  • Agal, S., Raulji, K., & Odedra, N. D. (2025). A machine learning approach to risk based asset allocation in portfolio optimization. Scientific Reports.
  • Babiak, M., & Baruník, J. (2026). Deep learning, predictability, and optimal portfolio returns. Journal of Empirical Finance.
  • Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research. Journal of Personality and Social Psychology, 51(6), 1173-1182.
  • Betancourt, C., & Chen, W.-H. (2021). Deep reinforcement learning for portfolio management of markets with a dynamic number of assets. Expert Systems with Applications.
  • Black, F., & Litterman, R. (1992). Global portfolio optimization. Financial Analysts Journal.
  • Board of Governors of the Federal Reserve System. (2026). 3-month Treasury bill secondary market rate, discount basis [RIFSGFSM03NA]. FRED, Federal Reserve Bank of St. Louis.
  • DeMiguel, V., Garlappi, L., & Uppal, R. (2009). Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy? The Review of Financial Studies.
  • Freyberger, J., Neuhierl, A., & Weber, M. (2020). Dissecting characteristics nonparametrically. The Review of Financial Studies.
  • Gollart, M., & Okhrin, Y. (2025). A reinforcement learning approach to dynamic portfolio optimization. Annals of Operations Research.
  • Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. The Review of Financial Studies.
  • Ha, Y. S., Kang, J., Kim, J., Chun, D., et al. (2026). Machine learning-based portfolio optimization: Comparative analysis with the All-Weather portfolio strategy. Financial Innovation.
  • Heaton, J. B., Polson, N. G., & Witte, J. H. (2016). Deep learning in finance. arXiv.
  • Jang, J., & Seong, N. (2023). Deep reinforcement learning for stock portfolio optimization by connecting with modern portfolio theory. Expert Systems with Applications.
  • Jiang, Z., Xu, D., & Liang, J. (2017). A deep reinforcement learning framework for the financial portfolio management problem. arXiv.
  • Kelly, B., Pruitt, S., & Su, Y. (2019). Characteristics are covariances: A unified model of risk and return. Journal of Financial Economics.
  • Kozak, S., Nagel, S., & Santosh, S. (2020). Shrinking the cross-section. Journal of Financial Economics.
  • Ledoit, O., & Wolf, M. (2004). Honey, I shrunk the sample covariance matrix. Journal of Portfolio Management.
  • López de Prado, M., Simonian, J., Fabozzi, F. A., & Fabozzi, F. J. (2025). Enhancing Markowitz's portfolio selection paradigm with machine learning. Annals of Operations Research.
  • Ma, Y., Han, R., & Wang, W. (2021). Portfolio optimization with return prediction using deep learning and machine learning. Expert Systems with Applications.
  • Markowitz, H. (1952). Portfolio selection. The Journal of Finance.
  • Michaud, R. O. (1989). The Markowitz optimization enigma: Is 'optimized' optimal? Financial Analysts Journal.
  • Ngo, V. M., Nguyen, H. H., & Nguyen, P. V. (2023). Does reinforcement learning outperform deep learning and traditional portfolio optimization models in frontier and developed financial markets? Research in International Business and Finance.
  • Sharpe, W. F. (1966). Mutual fund performance. The Journal of Business, 39(1), 119-138.
  • Simon, F., Weibels, S., & Zimmermann, T. (2026). Deep parametric portfolio policies. Management Science.
  • Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology, 13, 290-312.
  • Sortino, F. A., & Price, L. N. (1994). Performance measurement in a downside risk framework. The Journal of Investing, 3(3), 59-64.
  • Wang, W., Li, W., Zhang, N., & Liu, K. (2020). Portfolio formation with preselection using deep learning from long-term financial data. Expert Systems with Applications.
  • Wang, Y., Gao, H., Harvey, C. R., Liu, Y., & Tao, X. (2026). Machine learning meets Markowitz (NBER Working Paper No. 34861). National Bureau of Economic Research.
  • Zhang, Z., Zohren, S., & Roberts, S. (2020). Deep learning for portfolio optimization. arXiv.

Share your valuable work from Social Media Buttons