การเปรียบเทียบตัวแบบพยากรณ์ราคาส้มเขียวหวานโดยวิธีการปรับเรียบด้วยเส้นโค้งเลขชี้กำลัง

Abstract

Abstract

The objective of this study was to compare the tangerine prices forecast model by five exponential smoothing methods, i.e. 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. Time series of monthly tangerine prices gathered from the website of Office of Agricultural Economics during January, 2005 to October, 2017 of 154 values were divided into 2 sets. The first set had 144 values from January, 2005 to December, 2016 for constructing the forecasting models. The second set had 10 values from January to October, 2017 for comparing 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 damped trend exponential smoothing method and the forecasting model was where m represents the number of periods to forecast ahead with the starting value of January, 2017 (m = 1). 

Keywords: tangerine; exponential smoothing method; mean absolute percentage error (MAPE); root mean squared error (RMSE)