The Comparative Study on the Time Series Forecasting by Exponential Smoothing Techniques and Box-Jenkins Techniques : A Case Study of Forecasting the Export Values of Rice, Rubber and Cassava

Abstract

The purpose of this research is to compare between two forecasting techniques,
Exponential Smoothing Techniques and Box-Jenkins Techniques. In the study, the top three of Thai
Agricultural Export Values, rice, rubber and cassava, were analyzed by the both techniques. The
results were found that, the ARIMA model by Box-Jenkins Techniques is the best model which has
the minimum error. Also, the error (from the ARIMA models) is corresponding with theoretical
assumptions rather than the Exponential Smoothing models.