The renowned Indonesian company, PT Gojek Tokopedia Tbk, has a significant impact on the Indonesian economy by attracting investors to invest their shares. This study uses stock closing price data to forecast stock prices using ARIMA (AutoRegressive Intergrated Moving Average) and LSTM (Long Short-Term Memory) models, to predict using prediction by dividing the data into groups of 10 or 20 data with data sets to be trained as multiples. The analysis shows that ARIMA is superior to LSTM based on the comparison of average error and average percentage error, where the average error results in LSTM (3.843) and ARIMA (3.259), as well as the average error of LSTM (4.04%) and ARIMA (3.57%). The research supports the conclusion that ARIMA has a better performance in predicting the stock price of PT Gojek Tokopedia Tbk. These results provide important insights for investors and market participants, while the research supports the increased use of seasonal patterns in ARIMA forecasting for more accurate results in the future. Future research is recommended to explore additional factors and optimized models to further improve stock price prediction.
Copyrights © 2024