MAPE accuracy of CPO Forecasting by Applying Fuzzy Time Series

Abstract

Accuracy in each forecasting is needed in order to find out how much error will occur from the forecasting results. The forecasting process is included in the data mining process, which is to convert data into new knowledge. Forecasting is also widely used by commodity business players such as CPO (Crude Palm Oil). CPO is a staple of human life so there is a need for a technique to increase the business. one of the things that can be forecasted from CPO is the price of CPO. By using the times series, it is possible to predict CPO prices. The accuracy measurement that can be used is MAPE (Mean Absolute Percentage Error) for forecasting. In the test, the time series data used is the average monthly yield of CPO prices starting from December 2010 to April 2021 as many as 125 data. In this article, we tested 60% of the training data and 40% of the test data, and the MAPE obtained was 0.03463929%, and the MAPE obtained with 80% of the training data and 20% of the test data was 0.04392103%. The result of MAPE does not seem that much different so that it can be tested in various other accuracy techniques