espEconomy: strategy and practiceEconomy: strategy and practice1997-99672663-550XИнститут экономики1997-9967-2026-3-89-103esp-2198Research ArticleИННОВАЦИИ И ЦИФРОВАЯ ЭКОНОМИКАINNOVATION AND THE DIGITAL ECONOMYПрогнозирование динамики фондового рынка Туниса с использованием нейронных сетейForecasting Tunisian Stock Market Dynamics Using Neural Networkshttps://orcid.org/0009-0009-9913-8571БуфаресЮ.BoufaresYo.

Буфарес Ю. – PhD докторант, факультет экономики и менеджмента Сусса

Сусс 

Youssef Boufares – PhD student, Researcher, , Faculty of Economics and Management

Sousse 

youssef.bouf@gmail.comСмидаМ.SmidaM.

Смида М. – профессор, PhD, руководитель исследовательской лаборатории, факультет экономики и менеджмента Сусса

Сусс 

Mounir Smida – Professor, PhD, Director of Research Laboratory, Faculty of Economics and Management

Sousse 

mounirsmida2006@gmail.comНаучно-исследовательская лаборатория MOFID; Университет СуссаТунисMOFID Research Laboratory; University of SousseTunisiaУниверситет СуссаТунисUniversity of SousseTunisia20261610202621389103Copyright © Буфарес Ю., Смида М., 20262026Буфарес Ю., Смида М.Boufares Y., Smida M.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://esp.ieconom.kz/jour/article/view/2198

Прогнозирование динамики фондового рынка остается сложной задачей, особенно для развивающихся и пограничных рынков, где высокая неопределенность, ограниченная ликвидность и информационные асимметрии могут снижать устойчивость прогнозных моделей. Цель данного исследования заключается в сравнении прогностической эффективности моделей искусственных нейронных сетей и долгой краткосрочной памяти при прогнозировании индекса TUNINDEX20 с наивной моделью персистентности. Информационную базу составили ежедневные данные индекса TUNINDEX20 Тунисской фондовой биржи за период с 6 февраля 2012 г. по 31 декабря 2024 г., включающие 1 915 наблюдений. Выборка была разделена в хронологическом порядке на обучающую часть из 1 532 наблюдений и тестовую часть из 383 наблюдений. Результаты показали, что искусственная нейронная сеть несколько превосходит модель долгой краткосрочной памяти по основным показателям точности: для первой модели MAE составил 12.06, RMSE – 15.92, MAPE – 0.31%, MASE – 1.0769, R² – 0.9912, а точность прогнозирования направления движения индекса – 53.91%. Вместе с тем наивная модель показала немного более низкие ошибки прогноза (MAE = 11.76; RMSE = 15.82), а тест Диболда–Мариано не выявил статистически значимого преимущества ANN над наивной моделью (p = 0.7285). Полученные результаты показывают, что что более сложная архитектура не обеспечивает автоматического преимущества в прогнозировании TUNINDEX20. Перспективы дальнейших исследований связаны с включением технических индикаторов и макрофинансовых переменных, тестированием моделей на других развивающихся рынках, а также применением адаптивных и гибридных методов прогнозирования.

Predicting stock market dynamics remains challenging, especially in emerging and frontier markets, where high uncertainty, limited liquidity, and information asymmetries can reduce the robustness of predictive models. This study aims to compare the forecasting performance of Artificial Neural Networks (ANN) and Long Short-Term Memory (LSTM) models for the TUNINDEX20 against a naïve persistence benchmark. The information base consisted of daily data on the TUNINDEX20 index of the Tunis Stock Exchange from February 6, 2012, to December 31, 2024, including 1,915 observations. The sample was divided chronologically into a training set of 1,532 observations and a test set of 383 observations. The results showed that the artificial neural network slightly outperformed the long short‑term memory model in terms of key accuracy metrics: for the first model, MAE was 12.06, RMSE was 15.92, MAPE was 0.31%, MASE was 1.0769, R² was 0.9912, and the accuracy of predicting the direction of index movement was 53.91%. At the same time, the naive model showed slightly lower forecast errors (MAE = 11.76; RMSE = 15.82), and the Diebold–Mariano test did not reveal a statistically significant advantage of ANN over the naive model (p = 0.7285). The results obtained show that a more complex architecture does not automatically provide an advantage in forecasting TUNINDEX20. Future research could include adding technical indicators and macro‑financial variables, testing models in other emerging markets, and applying adaptive and hybrid forecasting methods.

фондовый рынокнейронная сетьцифровизацияинвестиционная стратегияфинансовое решениефинансовое прогнозированиеТунисStock MarketNeural NetworkDigitalizationInvestment StrategyFinancial DecisionFinancial ForecastingTunisiaReferencesBachelier, L. (1900). Théorie de la spéculation. Annales Scientifiques de l’École Normale Supérieure, 17, 21–86. https://doi.org/10.24033/asens.476Bachelier, L. (1900). Théorie de la spéculation. Annales Scientifiques de l’École Normale Supérieure, 17, 21–86. https://doi.org/10.24033/asens.476Bekaert, G., & Harvey, C. R. (1997). Emerging equity market volatility. Journal of Financial Economics, 43(1), 29–77. https://doi.org/10.1016/S0304-405X(96)00889-6Bekaert, G., & Harvey, C. R. (1997). Emerging equity market volatility. Journal of Financial Economics, 43(1), 29–77. https://doi.org/10.1016/S0304-405X(96)00889-6Bekaert, G., Harvey, C. R., & Mondino, T. (2023). Emerging equity markets in a globalised world. Emerging Markets Review, 56,101034. https://doi.org/10.1016/j. ememar.2023.101034Bekaert, G., Harvey, C. R., & Mondino, T. (2023). Emerging equity markets in a globalised world. Emerging Markets Review, 56,101034. https://doi.org/10.1016/j. ememar.2023.101034BVMT. (2025). Tunis Stock Exchange. Retrieved June 20, 2025 from https://tunis-stockexchange.com/BVMT. (2025). Tunis Stock Exchange. Retrieved June 20, 2025 from https://tunis-stockexchange.com/Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(3), 307–327. https://doi.org/10.1016/0304- 4076(86)90063-1Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity. Journal of Econometrics, 31(3), 307–327. https://doi.org/10.1016/0304- 4076(86)90063-1Boubaker, H., Saidane, B., & Ben Saad Zorgati, M. (2022). Modelling the dynamics of stock market in the Gulf Cooperation Council countries: Evidence on persistence to shocks. Financial Innovation, 8, Article 46. https://doi.org/10.1186/s40854-022-00348-3Boubaker, H., Saidane, B., & Ben Saad Zorgati, M. (2022). Modelling the dynamics of stock market in the Gulf Cooperation Council countries: Evidence on persistence to shocks. Financial Innovation, 8, Article 46. https://doi.org/10.1186/s40854-022-00348-3Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2008). Time series analysis: Forecasting and control (4th ed.). John Wiley & Sons. https://doi. org/10.1002/9781118619193Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2008). Time series analysis: Forecasting and control (4th ed.). John Wiley & Sons. https://doi. org/10.1002/9781118619193Büyükşahin, Ü.Ç., & Ertekin, Ş. (2019). Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition. Neurocomputing, 361, 151-163. https://doi. org/10.1016/j.neucom.2019.05.099Büyükşahin, Ü.Ç., & Ertekin, Ş. (2019). Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition. Neurocomputing, 361, 151-163. https://doi. org/10.1016/j.neucom.2019.05.099Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals, and Systems, 2(4), 303–314. https://doi. org/10.1007/BF02551274Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals, and Systems, 2(4), 303–314. https://doi. org/10.1007/BF02551274Fakhfekh, M., Jeribi, A., & Ben Salem, M. (2021). Volatility dynamics of the Tunisian stock market before and during the COVID‐19 outbreak: Evidence from the GARCH family models. International Journal of Finance & Economics, 28(2), 1653-1666. https://doi.org/10.1002/ ijfe.2499Fakhfekh, M., Jeribi, A., & Ben Salem, M. (2021). Volatility dynamics of the Tunisian stock market before and during the COVID‐19 outbreak: Evidence from the GARCH family models. International Journal of Finance & Economics, 28(2), 1653-1666. https://doi.org/10.1002/ ijfe.2499Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. The Journal of Finance, 25(2), 383–417. https://doi.org/10.2307/2325486Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. The Journal of Finance, 25(2), 383–417. https://doi.org/10.2307/2325486Fischer, T., & Krauss, C. (2018). Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2), 654–669. https://doi.org/10.1016/j. ejor.2017.11.054Fischer, T., & Krauss, C. (2018). Deep learning with long short-term memory networks for financial market predictions. European Journal of Operational Research, 270(2), 654–669. https://doi.org/10.1016/j. ejor.2017.11.054Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. Review of Financial Studies, 33(5), 2223–2273. https://doi.org/10.1093/rfs/ hhaa009Gu, S., Kelly, B., & Xiu, D. (2020). Empirical asset pricing via machine learning. Review of Financial Studies, 33(5), 2223–2273. https://doi.org/10.1093/rfs/ hhaa009Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Literature review: Machine learning techniques applied to financial market prediction. Expert Systems with Applications, 124, 226–251. https://doi.org/10.1016/j. eswa.2019.01.012Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Literature review: Machine learning techniques applied to financial market prediction. Expert Systems with Applications, 124, 226–251. https://doi.org/10.1016/j. eswa.2019.01.012Hochreiter, S., & Schmidhuber, J. (1997). Long shortterm memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735Hochreiter, S., & Schmidhuber, J. (1997). Long shortterm memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359–366. https://doi. org/10.1016/0893-6080(89)90020-8Hornik, K., Stinchcombe, M., & White, H. (1989). Multilayer feedforward networks are universal approximators. Neural Networks, 2(5), 359–366. https://doi. org/10.1016/0893-6080(89)90020-8Kaygın, C. Y., Gün, M., Akarsu, O. N., Bağcı, H., & Yanık, A. (2026). Algorithmic stability in turbulent markets: Unveiling the superiority of shallow learning over deep architectures in cryptocurrency forecasting. Mathematics, 14(6),989. https://doi.org/10.3390/math14060989Kaygın, C. Y., Gün, M., Akarsu, O. N., Bağcı, H., & Yanık, A. (2026). Algorithmic stability in turbulent markets: Unveiling the superiority of shallow learning over deep architectures in cryptocurrency forecasting. Mathematics, 14(6),989. https://doi.org/10.3390/math14060989Mehtab, S., & Sen, J. (2020). Stock price prediction using CNN and LSTM-based deep learning models. In Proceedings of the 2020 International Conference on Decision Aid Sciences and Application (DASA) (pp. 448–453). IEEE. https://doi.org/10.1109/ DASA51403.2020.9317207Mehtab, S., & Sen, J. (2020). Stock price prediction using CNN and LSTM-based deep learning models. In Proceedings of the 2020 International Conference on Decision Aid Sciences and Application (DASA) (pp. 448–453). IEEE. https://doi.org/10.1109/ DASA51403.2020.9317207Nayak, R. K., Misra, D., & Behera, H. S. (2014). Impact of data normalization on stock index forecasting. International Journal of Computer Information Systems and Industrial Management Applications, 6, 257–269.Nayak, R. K., Misra, D., & Behera, H. S. (2014). Impact of data normalization on stock index forecasting. International Journal of Computer Information Systems and Industrial Management Applications, 6, 257–269.Samuelson, P. A. (1965). Proof that properly anticipated prices fluctuate randomly. Industrial Management Review, 6(2), 41–49.Samuelson, P. A. (1965). Proof that properly anticipated prices fluctuate randomly. Industrial Management Review, 6(2), 41–49.Sezer, O. B., Gudelek, M. U., & Ozbayoglu, A. M. (2020). Financial time series forecasting with deep learning: A systematic literature review (2005–2019). Applied Soft Computing, 90, 106181. https://doi.org/10.1016/j. asoc.2020.106181Sezer, O. B., Gudelek, M. U., & Ozbayoglu, A. M. (2020). Financial time series forecasting with deep learning: A systematic literature review (2005–2019). Applied Soft Computing, 90, 106181. https://doi.org/10.1016/j. asoc.2020.106181Shahrour, M., & Dekmak, M. (2024). Intelligent stock prediction: A neural network approach. International Journal of Financial Engineering, 11(1), 2250016. https://doi.org/10.1142/S2424786322500165Shahrour, M., & Dekmak, M. (2024). Intelligent stock prediction: A neural network approach. International Journal of Financial Engineering, 11(1), 2250016. https://doi.org/10.1142/S2424786322500165Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(56), 1929–1958. https:// doi.org/10.5555/2627435.2670313Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(56), 1929–1958. https:// doi.org/10.5555/2627435.2670313Tashman, L. J. (2000). Out-of-sample tests of forecasting accuracy: An analysis and review. International Journal of Forecasting, 16(4), 437–450. https://doi. org/10.1016/S0169-2070(00)00065-0Tashman, L. J. (2000). Out-of-sample tests of forecasting accuracy: An analysis and review. International Journal of Forecasting, 16(4), 437–450. https://doi. org/10.1016/S0169-2070(00)00065-0Tsay, R. S. (2005). Analysis of financial time series (2nd ed.). John Wiley & Sons. https://doi. org/10.1002/0471746193Tsay, R. S. (2005). Analysis of financial time series (2nd ed.). John Wiley & Sons. https://doi. org/10.1002/0471746193Varshney, S., & Srivastava, P. (2023). A comparative study of future stock price prediction through artificial neural network and ARIMA modelling. NMIMS Management Review, 31(4), 229–239. https://doi. org/10.1177/09711023241230367Varshney, S., & Srivastava, P. (2023). A comparative study of future stock price prediction through artificial neural network and ARIMA modelling. NMIMS Management Review, 31(4), 229–239. https://doi. org/10.1177/09711023241230367Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14(1), 35–62. https://doi.org/10.1016/S0169-2070(97)00044-7Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with artificial neural networks: The state of the art. International Journal of Forecasting, 14(1), 35–62. https://doi.org/10.1016/S0169-2070(97)00044-7

The authors declare that there are no conflicts of interest present.