BCA Stock Price Prediction Using Time Series Method With GRU (Gated Recurrent Unit)

Abstract

Stock price prediction is a crucial component in investment decision-making, enabling investors to plan strategies more accurately and minimize risks. This study applies the Gated Recurrent Unit (GRU) model to predict the stock prices of blue-chip banking companies in Indonesia using data from the period 2019 to 2024. The model utilizes historical stock data to forecast future trends. The results from the first testing scheme, with a data split ratio of 70% / 30%, using GRU units (128,256) with the Adam optimizer, show that the GRU model is the most optimal in terms of prediction, measured by metrics such as MSE, RMSE, and MAPE. This study also proposes a web-based dashboard that visualizes the predicted stock prices and provides decision-support tools for investors. The findings highlight the effectiveness of deep learning in financial forecasting and underscore its potential to enhance investment strategies.