Comparison of Tapioca Selling Prices Forecast Model by Statistical Forecasting Methods

Abstract

The objective of this study was to compare the tapioca prices forecast model by six statistical forecasting methods: Box-Jenkins method, Holt’s exponential smoothing method, damped trend exponential smoothing method, simple seasonal exponential smoothing method, Winters’ additive exponential smoothing method, and Winters’ multiplicative exponential smoothing method in order to create the best forecasting model. Time series of monthly tapioca prices gathered from the website of Bank of Thailand during January, 1999 to September, 2018 of 237 values were divided into two sets. The first set had 235 values from January, 1999 to July, 2018 for constructing the forecasting models. The second set had two values from August to September, 2018 for comparing the accuracy of the forecasts via the criteria of the lowest mean absolute percentage error and root mean squared error. Research findings indicated that for all forecasting methods that had been studied, the most accurate method was Winters’ additive exponential smoothing method.