ตัวแบบพยากรณ์มูลค่าการส่งออกยางพาราของประเทศไทย

Abstract

The objectives of this research were to study and compare forecasting methods for forecasting the export value of rubber of Thailand. The data gathered from the website of Rubber Intelligence Unit during January 2013 to September 2019. The 81 values were used and divided into 2 sets. The first set had 72 values from January 2013 to December 2018 for constructing the forecasting models by the most suitable three methods to this time series which were double moving average method, double exponential smoothing method, and Box–Jenkins method. The second set had 9 values from January 2019 to September 2019 for comparing accuracy of the forecast via the criteria of the lowest mean absolute deviation, mean square error, and mean absolute percentage error. Research findings indicated that the Box-Jenkins method was the most accurate.