Forecasting Model for the Prices of Assorted Mangosteen

Abstract

The purpose of this research was to construct the suitable forecasting model for the prices of assorted mangosteen. The data gathered from the website of Office of Agricultural Economics during January, 2005 to August, 2015 (128 values) was used and divided into two categories. The first category had 120 values, which were the data taken from January, 2005 to December, 2014 for fitting the model by the methods of Box-Jenkins, decomposition, and Winters’ multiplicative exponential smoothing. The second category had 8 values which were the data collected from January to August, 2015, for checking the accuracy of the forecasting models via the criterion of the minimum root mean squared error. The results showed that the decomposition method was the most suitable for this time series among all the studied forecasting methods.