Hybrid Artificial Intelligence and Traditional Models for Stock Price Forecasting: Evidence from Selected Banks Listed on the Iraq Stock Exchange 2010–2024

Ibrahim Baraka Hamadi

doi:10.36406/ijbam.v9i1.352

Authors

Ibrahim Baraka Hamadi College of Administration and Economics, University of Maysan, Iraq. https://orcid.org/0009-0000-0817-5460

DOI:

https://doi.org/10.36406/ijbam.v9i1.352

Keywords:

Hybrid Models, Forecasting, LSTM, ARIMA, Emerging Markets

Abstract

Purpose- This study aims to develop and evaluate a hybrid approach for forecasting the stock prices of selected banks listed on the Iraq Stock Exchange (ISX) by integrating traditional statistical analysis using autoregressive integrated moving average (ARIMA) with artificial intelligence using long short-term memory (LSTM).

Design/methodology/approach- The study uses annual data from Iraqi banks covering the period 2010–2024, including annual closing stock prices, return on assets (ROA), return on equity (ROE), and earnings per share (EPS). The data were divided into training data for 2010–2020 and testing data for 2021–2024. The LSTM model consisted of two layers with 64 units and a dropout rate of 0.2 and was trained using the Adam optimizer. The predictive performance of the hybrid ARIMA–LSTM model was evaluated using mean absolute percentage error (MAPE), root mean square error (RMSE), and the coefficient of determination (R²).

Findings- The results indicate that the hybrid ARIMA–LSTM model outperformed the standalone classical models in terms of forecasting accuracy, achieving the lowest MAPE and RMSE values and an R² of 0.99. These findings demonstrate the effectiveness of integrating linear and nonlinear modeling approaches for stock price forecasting. Furthermore, EPS was identified as the most strongly associated factor with stock price movements, with correlation coefficients ranging from 0.956 to 0.985.

Originality/value- This study contributes to the literature by applying a hybrid ARIMA–LSTM framework to stock price forecasting in the Iraqi banking sector, an emerging market context that remains relatively underexplored. The findings highlight the potential of combining traditional time-series models with artificial intelligence techniques to improve forecasting accuracy and provide additional insights into the role of profitability indicators in stock price movements.

Downloads

Download data is not yet available.

References

Al-Rashidi, H. A. H. J. (2023). Al-namādhij al-khaṭṭiyyah wa ghayr al-khaṭṭiyyah li-gharaḍ al-tanabbu' bi-sūq al-awrāq al-māliyyah al-'irāqīyah [Linear and nonlinear models for forecasting the Iraqi stock market] [Doctoral dissertation, University of Al-Qadisiyah].

Baghdad Bank. (2024). Al-bayānāt al-māliyyah al-marḥaliyyah [Interim financial statements].

Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (2016). Time series analysis: Forecasting and control (5th ed.). Wiley.

Central Bank of Iraq. (2023). Al-taqrīr al-iqtiṣādī al-sanawī [Annual economic report].

Central Bank of Iraq. (2024). Al-khuṭṭah al-istrātījiyyah al-thālithah 2024–2026 [Third strategic plan 2024–2026].

Choudhury, M. M. (2024). Signaling theory: An approach to organizational behavior research. Journal of Accounting, Business and Management, 31(2), 98–120. https://doi.org/10.31966/jabminternational.v31i2.1199

Hamiane, S., Youssef, G., Hamid, K., & Meryam, T. (2024). Comparative analysis of LSTM, ARIMA, and hybrid models. Ingénierie des Systèmes d'Information, 29(3). https://doi.org/10.18280/isi.290306

Iraqi Securities Commission. (2023). Al-taqrīr al-sanawī li-adā' al-sūq al-mālī al-'irāqī [Annual report on the performance of the Iraqi financial market].

Luo, Z., Wei, G., Qingfu, L., & Zhengjun, Z. (2020). A hybrid model for financial time-series forecasting. Expert Systems, 38(2), e12633. https://doi.org/10.1111/exsy.12633

Rather, A. M., Agarwal, A., & Sastry, V. N. (2015). Recurrent neural network and a hybrid model for prediction of stock returns. Expert Systems with Applications, 42(6), 3234–3241. https://doi.org/10.1016/j.eswa.2014.12.003

Saâdaoui, F., & Rabbouch, H. (2024). Financial forecasting improvement with LSTM-ARFIMA hybrid models and non-Gaussian distributions. Technological Forecasting and Social Change, 206, 123539. https://doi.org/10.1016/j.techfore.2024.123539

Shi, Z., Hu, Y., Mo, G., & Wu, J. (2022). Attention-based CNN-LSTM and XGBoost hybrid model. arXiv. https://doi.org/10.48550/arXiv.2204.02623

Spence, M. (1973). Job market signaling. The Quarterly Journal of Economics, 87(3), 355–374. https://doi.org/10.2307/1882010

Zhang, G. P. (2003). Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 50, 159–175. https://doi.org/10.1016/S0925-2312(01)00702-0

Zhang, J., Liu, H., Bai, W., & Li, X. (2024). A hybrid approach of wavelet transform, ARIMA and LSTM model for the share price index futures forecasting. The North American Journal of Economics and Finance, 69, 102022. https://doi.org/10.1016/j.najef.2023.102022