<![CDATA[Stock price prediction is one of the complex problems in the financial world because it is non-linear and influenced by many external factors. This study aims to apply the Long Short-Term Memory (LSTM) algorithm, a variant of Recurrent Neural Network (RNN) within the Deep Learning framework, to predict the closing price of banking sector stocks in the Indonesian capital market. The data used comes from five major banking issuers on the Indonesia Stock Exchange, namely BBCA, BBTN, BBRI, BBMRI, and BBNI, obtained from the Kaggle platform. The method used is a literature study by reviewing relevant literature related to the application of LSTM to financial time series data. The main parameter configurations include 50 hidden units, 100 epochs, a batch size of 32, a learning rate of 0.001, and a ReLU activation function with Adam optimizer. Model evaluation was carried out using the Root Mean Square Error (RMSE) metric of 6.61%, Mean Absolute Error (MAE) of 1.74%, Mean Square Error (MSE) of 43.69%, and a coefficient of determination R² of 0.718. The results show that the LSTM model is able to produce predictions close to actual values, reduce the risk of overfitting, and can be used as an investment decision-making tool for investors in the Indonesian stock market.]]>
Copyrights © 2026
