Comparison between box-jenkins and fuzzy time series to predict crude palm oil production / Nurul Najiehah Azamil

In Malaysia, the palm oil industry has been an important agriculture sector that gives great impact on Malaysia’s economy. The increment in palm oil’s demand gives an impact on countries that produce and exported palm oil especially in Malaysia and Indonesia. This study has been conducted to forecast the production of palm oil product (crude palm oil) in 2021. The secondary data of crude palm oil production were used in this study in order to fulfill the objective which are to generate a model of crude palm oil production in Malaysia using the Box-Jenkins method and Fuzzy Time Series method in order to know the monthly production pattern of crude palm oil from January 2013 to December 2019. Other than that, it is to identify the best forecasting model between Box-Jenkins method and Fuzzy Time Series by looking at the lowest MSE and RMSE values of each model and to forecast the production of crude palm oil in Malaysia for the year 2021 based on the selected best model. Moreover, Box-Jenkins method was generated through R-Studio package software and Fuzzy Time Series method was generated by using Microsoft Excel. The error measure was calculated in this study which are Mean Square Error (MSE) and Root Mean Square Error (RMSE) in making comparison on which methods have the lowest value. The result shows that Box-Jenkins method has the lowest MSE value which is 11025441000 compared to the Fuzzy Time Series method that is 11307681646.807. For RMSE value, the Box-Jenkins is 105002.1 while Fuzzy Time Series is 106337.58. Thus, the best forecasting model that has the lowest MSE and RMSE values in this study to forecast the production of crude palm oil in Malaysia for the year 2021 is the Box-Jenkins method.