Abstract
The evolution of global financial markets has heightened the demand for accurate stock price prediction methods, particularly in the Islamic banking sector, which operates under unique principles of Sharia compliance. This study aims to predict the stock prices of Bank Syariah Indonesia (BSI) using a Bidirectional Long Short-Term Memory (Bi-LSTM) model. The dataset comprises daily closing prices from January 2022 to June 2024. The model is optimized through systematic hyperparameter tuning, including configurations for the number of layers, neurons, batch size, learning rate, and optimizers. Evaluation using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) identifies the Adam optimizer with a learning rate of 0.001 and batch size of 16 as the optimal configuration. The results highlight that while increasing the number of neurons or layers reduces minimum error, it increases model instability. This research provides novel insights into the application of Bi-LSTM for predicting Islamic banking stock prices, supporting data-driven decision-making in the Islamic financial sector.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call