Price prediction of the Borsa Istanbul banks index with traditional methods and artificial neural networks

Abstract

In terms of asset size, the banking system constitutes 83% of the financial markets in Turkiye. Given the importance of the banking system in the Turkish capital market, this study offers a price forecasting analysis of the Borsa Istanbul Banks Index, which represents the domestic banking system, between December 27, 1996, to August 31, 2023, using the traditional Autoregressive Integrated Moving Average (ARIMA) Model and two artificial intelligence-based deep learning models, namely, the Facebook Prophet Model (FPM) and Convolutional Neural Networks Model (CNNM). The findings indicate that the CNNM perform better than the other models. The results are useful for researchers working with time series data at the stage of method selection and investment firms and managers that are forecasting future stock price movements. Policy implications of the findings are discussed.