<![CDATA[Fuzzy time series is a forecasting method that handles data uncertainty by applying fuzzy set theory. This study compares the forecasting accuracy of the Chen fuzzy time series method, modified using Fuzzy C-Means (FCM), and the Lee method in predicting Indonesian coal prices from January 2019 to December 2023. The Chen method is en hanced by generating fuzzy intervals through FCM to better reflect data distribution, while the Lee method uses weighted fuzzy logical relationships. Forecast accuracy is measured using Mean Absolute Percentage Error (MAPE). The modified Chen method achieves a MAPE of 2.56% compared to 5.07% for the Lee method. These results show that cluster ing techniques like FCM can improve fuzzy time series forecasting. This contrasts with earlier studies that favored the Lee method and highlights the potential of adaptive interval construction for volatile commodity prices. The proposed modification offers a promising alternative for improving prediction accuracy in economic time series.]]>
Copyrights © 2025
