Forecasting model for spice export volume in Thailand

Abstract

The purpose of this study is to create a suitable forecasting model of the spice export volume for Thailand by 4 statistical methods: Holt’s exponential smoothing method, Brown’s exponential smoothing method, Box-Jenkins method, and combined forecast method approach using the ordinary least squares method to estimate the regression coefficient. Time series of monthly spice export volume (which were gathered from the website of the office of agricultural economics from January 2017 to December 2022 comprising 72 observations) were divided into two datasets. The first dataset had 66 observations from January 2017 to June 2022. The second dataset had 6 observations from July 2022 to December 2022. The first dataset was used for creating the forecasting model. The second dataset was used for checking the accuracy of the forecasting models via the criterion of the lowest of mean absolute percentage error and the lowest root mean square error. The results showed that Brown’s exponential Method is most suitable for this time series (MAPE = 28.93, RMSE = 222,979.17).