<![CDATA[Predicting stock prices plays a pivotal role in the decision-making processes of organizations and individual investors. This research focuses on the predicting weekly closing stock prices, specifically for UNTR, using the ARFIMA method. The ARFIMA method shows promise in handling long-memory data, but its effectiveness in predicting UNTR's stock prices requires thorough examination to ensure its applicability and reliability. The aim of this study is to predict the weekly closing prices of UNTR stocks using the ARFIMA method. The training data used spans from January 1, 2020, to December 31, 2022, with the objective of predicting the period from January 1, 2023, to February 28, 2023. The result shows that the ARFIMA (10; 0.4993; 3) model was selected due to its optimal performance, having the lowest RMSE and MAPE values, specifically an RMSE of 0.4 and a MAPE of 4.16%. This model successfully captures the long-term memory patterns in the data, generating accurate predictions for the projected period. ]]>
Copyrights © 2025
