Deep Learning versus Classical Machine Learning in Portfolio Optimization with Downside Risk: Evidence from the Abu Dhabi Securities Exchange (ADX)

Naji  Khaleef Ajeel; Ali  Souri; Ezatollah  Abbasian; Eslam  Fakher

Authors

Naji Khaleef Ajeel Ph.D. Student, Department of Finance – Financial Engineering, Aras International Campus, University of Tehran, Jolfa, Iran Author https://orcid.org/0009-0006-6684-0389
Ali Souri Associate Professor, Department of Economics, Faculty of Economics, University of Tehran, Tehran, Iran Author https://orcid.org/0000-0003-0630-3013
Ezatollah Abbasian Professor, Department of Financial Engineering, Faculty of Management, University of Tehran, Tehran, Iran https://orcid.org/0000-0001-8364-6461
Eslam Fakher Assistant Professor, Department of Financial Management, Shahid Chamran University of Ahvaz, Ahvaz, Iran Author https://orcid.org/0000-0002-5240-2795

Keywords:

Deep Learning; Portfolio Optimization; Mean Semi-Absolute Deviation; Downside Risk; DMLP; Support Vector Regression; Abu Dhabi Securities Exchange.

Abstract

This study aimed to develop and evaluate a hybrid portfolio optimization framework integrating deep learning-based stock-return forecasting with Mean Semi-Absolute Deviation (MSAD) optimization to improve risk-adjusted asset allocation in the Abu Dhabi Securities Exchange. An applied quantitative design was employed using daily Open, High, Low, Close, and trading-volume data for 25 active ADX stocks. Three deep learning architectures—Deep Multilayer Perceptron (DMLP), one-dimensional Convolutional Neural Network (1D-CNN), and Long Short-Term Memory (LSTM)—were compared with Support Vector Regression (SVR). Predictors were constructed from rolling 20-day windows, and the dataset was partitioned into training, validation, and out-of-sample testing subsets. Forecasts were incorporated into an MSAD linear optimization framework, and stocks were ranked into five quintile portfolios. Predictive performance was assessed using MAE, MSE, and directional Hit Rates, while portfolio performance was evaluated through daily return and return-to-risk ratios. DMLP produced the lowest forecasting errors, with MAE=0.0163 and MSE=0.0013, and achieved the highest positive directional accuracy (HR+=0.5163). In portfolio optimization, DMLP+MSAD generated the strongest results for aggressive portfolios, producing daily returns of 0.1817 and 0.2164 and return-to-risk ratios of 0.49 and 0.78 for P1 and P3, respectively. SVR+MSAD showed greater stability in the defensive P2 portfolio, whereas CNN+MSAD demonstrated comparative resilience in negatively performing portfolios. Across the quintiles, MSAD-optimized AI strategies consistently exhibited superior risk-adjusted performance relative to equal-weighted allocation. Integrating predictive artificial intelligence models with downside-risk-based MSAD optimization improves portfolio construction under asymmetric and volatile market conditions. DMLP+MSAD is particularly effective for high-return strategies, whereas SVR and CNN may offer advantages under more defensive or adverse market regimes, indicating that optimal model selection should reflect investors’ risk-return preferences.

References

[1] A. Frolov, R. Boiko, V. Rudevska, D. Butenko, and A. Moisiakha, "Evaluating Modern Quantitative Methods for Investment Portfolio Management under Market Uncertainty," Journal of Applied Economic Sciences, vol. 20, no. 3, 2025, doi: 10.57017/jaes.v20.3(89).05.

[2] D. Garg, "Multiscale Portfolio Optimization via ICEEMDAN and Hilbert Spectral Analysis," Research Square, 2025.

[3] B. Chen, J. Zhong, and Y. Chen, "A Hybrid Approach for Portfolio Selection with Higher-Order Moments: Empirical Evidence from Shanghai Stock Exchange," Expert Systems with Applications, vol. 145, p. 113104, 2020, doi: 10.1016/j.eswa.2019.113104.

[4] G. Nicolas, "Measuring Risks in a Portfolio of Financial Assets Using the Downside Risk Method," in Advances in Analysis, ed, 2020.

[5] J. Chen, E. Furman, S. Ricci, and J. Shanthirajah, "Mean-Tail Gini Framework for Optimal Portfolio Selection," arXiv, 2025. [Online]. Available: https://arxiv.org/abs/2509.17225

[6] R. Ghasemiyeh, H. Sinaee, and A. R. Dodangeh, "Portfolio Optimization Model Based on Forecasting Using CNN Neural Network and MSAD Metric in Tehran Stock Exchange," Strategic Research on Budgeting and Finance, vol. 5, no. 1, 2024.

[7] Y. Ma, R. Han, and W. Wang, "Prediction-Based Portfolio Optimization Models Using Deep Neural Networks," IEEE Access, vol. 8, pp. 115393-115405, 2020, doi: 10.1109/ACCESS.2020.3003819.

[8] W. Wang, W. Li, N. Zhang, and K. Liu, "Portfolio Formation with Preselection Using Deep Learning from Long-Term Financial Data," Expert Systems with Applications, vol. 143, p. 113042, 2020, doi: 10.1016/j.eswa.2019.113042.

[9] S. I. Lee and S. J. Yoo, "Threshold-Based Portfolio: The Role of the Threshold and Its Applications," The Journal of Supercomputing, vol. 76, no. 10, pp. 8040-8057, 2020, doi: 10.1007/s11227-018-2577-1.

[10] V. Piccialli and M. Sciandrone, "Nonlinear Optimization and Support Vector Machines," Annals of Operations Research, vol. 314, no. 1, pp. 15-47, 2022, doi: 10.1007/s10479-022-04655-x.

[11] W. C. Yeh, Y. Jiang, S. Y. Tan, and C. Y. Yeh, "A New Support Vector Machine Based on Convolution Product," Complexity, vol. 2021, no. 1, p. 9932292, 2021, doi: 10.1155/2021/9932292.

[12] M. R. Apriyadi and D. P. Rini, "Hyperparameter Optimization of Support Vector Regression Algorithm Using Metaheuristic Algorithm for Student Performance Prediction," International Journal of Advanced Computer Science and Applications, vol. 14, no. 2, 2023, doi: 10.14569/IJACSA.2023.0140218.

[13] F. D. Paiva, R. T. N. Cardoso, G. P. Hanaoka, and W. M. Duarte, "Decision-Making for Financial Trading: A Fusion Approach of Machine Learning and Portfolio Selection," Expert Systems with Applications, vol. 115, pp. 635-655, 2019, doi: 10.1016/j.eswa.2018.08.003.

[14] N. Bagheri Mazraeh, A. Daneshvar, and M. Madanchi Zaj, "Development of a New Ensemble Learning Approach for Stock Portfolio Selection Using Multiclass SVM and Genetic Algorithm," Financial Engineering and Portfolio Management, vol. 13, no. 50, pp. 282-305, 2022, doi: 10.1155/2022/5974842.

[15] M. Sarcheshmehpour, A. Khodamipour, M. Mohammadi, and H. Zeinali, "Application of Machine Learning Models in Optimal Stock Portfolio Construction and Comparison of Their Performance," Journal of Financial Engineering and Securities Management (Portfolio Management), vol. 11, no. 45, pp. 147-176, 2020.

[16] M. Vazan, Deep Learning: Principles, Concepts, and Approaches. Tehran: Mi'ad Andisheh Publications, 2020.

[17] O. B. Sezer, M. U. Gudelek, and A. M. Ozbayoglu, "Financial Time Series Forecasting with Deep Learning: A Systematic Literature Review: 2005-2019," Applied Soft Computing, vol. 90, p. 106181, 2020, doi: 10.1016/j.asoc.2020.106181.

[18] C. Zhang, N. N. A. Sjarif, and R. B. Ibrahim, "Deep Learning Techniques for Financial Time Series Forecasting: A Review of Recent Advancements: 2020-2022," arXiv, 2023. [Online]. Available: https://arxiv.org/abs/2305.04811

[19] M. Ahmadi, M. Khashei, and N. Bakhtiarvand, "Enhancing Air Quality Classification Using a Novel Discrete Learning-Based Multilayer Perceptron Model (DMLP)," International Journal of Environmental Science and Technology, vol. 22, no. 5, pp. 3051-3062, 2025, doi: 10.1007/s13762-024-06017-5.

[20] S. F. Memarzadeh, H. Khosravi Farsani, and T. Javdani Gandomani, "Presenting a Deep Learning-Based Method for Stock Price Prediction," in Proceedings of the 8th International Conference on Web Research (ICWR), Tehran, 2022.

[21] M. Heidari and H. Amiri, "Inspecting the Predictive Power of Artificial Intelligence Models in Predicting the Stock Price Trend in Tehran Stock Exchange," Financial Research Journal, vol. 24, no. 4, pp. 602-623, 2022.

[22] T. Fischer and C. Krauss, "Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions," European Journal of Operational Research, vol. 270, no. 2, pp. 654-669, 2018, doi: 10.1016/j.ejor.2017.11.054.

[23] J. Song, Q. Cheng, X. Bai, W. Jiang, and G. Su, "LSTM-Based Deep Learning Model for Financial Market Stock Price Prediction," Journal of Economic Theory and Business Management, vol. 1, no. 2, pp. 43-50, 2024.

[24] H. Kumar, A. Taluja, and P. Kumar, "A Comprehensive Analysis of LSTM Techniques for Predicting Financial Market," International Journal of Financial Engineering, vol. 11, no. 4, p. 2442004, 2024, doi: 10.1142/S2424786324420040.

[25] S. Zoghi, R. Raei, and S. Fallahpour, "Developing a Predictive Model for the Iranian Gold Coin Futures Market Using Long Short-Term Memory (LSTM) Algorithm," Journal of Financial Engineering and Securities Management (Portfolio Management), vol. 13, no. 53, pp. 34-53, 2022.

[26] V. D. Ta, C. M. Liu, and D. A. Tadesse, "Portfolio Optimization-Based Stock Prediction Using Long Short-Term Memory Network in Quantitative Trading," Applied Sciences, vol. 10, no. 2, p. 437, 2020, doi: 10.3390/app10020437.

[27] A. Nafia, A. Yousfi, and A. Echaoui, "Equity-Market-Neutral Strategy Portfolio Construction Using LSTM-Based Stock Prediction and Selection: An Application to S&P 500 Consumer Staples Stocks," International Journal of Financial Studies, vol. 11, no. 2, p. 57, 2023, doi: 10.3390/ijfs11020057.

[28] M. Zolfaghari, B. Sahabi, and M. J. Bakhtiari, "Designing a Model for Predicting the Total Stock Index Return (Emphasizing Hybrid Deep Learning Neural Network Models and GARCH Family Models)," Journal of Financial Engineering and Securities Management (Portfolio Management), vol. 11, no. 42, pp. 138-171, 2020.

[29] A. Sharif Far, M. Khaliliaraghi, I. Raeesi Vanani, and M. Fallahshams, "Application of Deep Learning Architectures in Stock Price Forecasting: A Convolutional Neural Network Approach," Journal of Asset Management and Financing, vol. 10, no. 3, pp. 1-20, 2022.

[30] S. Mehtab and J. Sen, "Analysis and Forecasting of Financial Time Series Using CNN and LSTM-Based Deep Learning Models," in Advances in Distributed Computing and Machine Learning: Proceedings of ICADCML 2021. Singapore: Springer Singapore, 2022, pp. 405-423.

[31] P. Barroso and K. Saxena, "Lest We Forget: Learn from Out-of-Sample Forecast Errors When Optimizing Portfolios," The Review of Financial Studies, vol. 35, no. 3, pp. 1222-1278, 2022, doi: 10.1093/rfs/hhab041.