Predicting ANTAM Stock Price with Deep Learning: A Tool for Strategic Investment Decisions

Abstract

In the dynamic and often unpredictable financial markets, accurate stock price predictions are crucial for making informed investment decisions. This study explores the application of deep learning techniques to predict the stock prices of PT Aneka Tambang (ANTAM), a prominent Indonesian mining company, with the aim of enhancing investment strategies. By utilizing historical stock price data, technical indicators, and relevant economic factors, we develop a comprehensive deep learning model to forecast future stock prices. The study employs deep learning architectures: Long Short-Term Memory (LSTM) networks capture both temporal dependencies and intricate patterns within the financial data. The findings indicate that deep learning models, particularly those leveraging complex feature sets, provide more accurate and timely predictions. The result of forecasting ANTAM stock price using an LSTM model, trained over 250 epochs, resulted in a Root Mean Square Error (RMSE) of 26,212, indicating a reasonable level of prediction accuracy. Furthermore, the research examines the practical implications of integrating deep learning predictions into investment decision-making processes. Based on the forecast model, the investment decision suggestion is to buy ANTAM stock.