A Model Comparison for Rubber Price Forecasting in Thailand
Article Sidebar
Main Article Content
Abstract
The objective of this research was to study and compare an appropriate model of rubber price forecasting. Three models of Holt’s method, the Double exponential smoothing method, and the Box–Jenkins method were represented. In this research, the rubber price data of Thailand was used. The total data was collected in 132 months between January 2011 to December 2021, and divided data into two sets. The first set had 120 months from January 2011 to December 2020 for constructing 3 models, and another set had 12 months (January 2021 to December 2021) for comparing accuracy of the forecasts via the criteria of the lowest mean absolute percentage error. It was concluded from this research that using Holt’s method for rubber price forecasting in Thailand which model was suitable while the defaults were
and
because the mean absolute percentage error was the lowest value (6.608).
Article Details
References
Bowerman, B. L., & O’Cannell R. T. (1993). Forecasting and time series: An applied approach. (The duxbury advanced series in statistics and decision sciences). Belmont, CA: Duxbury Press.
Box, G. E. P., Jenkins, G. M., & Reinsel, G. C. (1994). Time series analysis: Forecasting and control (3rd ed.). New Jersey: Prenntice Hall.
Cooperative fund rubber thongphaphum. (2013). Good quality raw rubber sheet. Accessed August 20, 2021. Retrieved from https://www.thansettakij.com/economy/489891
Keerativibool, W., (2015). Forecasting model for the price of rubber unsmoked sheets level 3. RMUTI Journal Science and Technology, 8(2), 131–139. (in Thai)
Keerativibool, W., & Kanjanasamranwong, P. (2014). Forecasting the prices of rubber smoked sheets level 3. KKU Science Journal, 42(1), 235 – 247. (in Thai)
Lorchirachoolnkul, V., & Jitthavech, J. (2005). Forecast techniques (3rd ed.). Bangkok: National institute of development administration (NIDA). (in Thai)
