The Forecasting Techniques Comparison of Field Crops in Thailand
Keywords:Field crops, Time-series forecasting technique, Mean absolute percentage error, Tracking signal
The essential agricultural production of field crops in Thailand was forecasted in this study. Time-series approaches were used to forecast six field crops, including Thai white rice 15%, Thai jasmine rice 105, sugar cane, cassava, peanuts, and corn, and the results were compared to the mean absolute percentage error and tracking signal. For forecasting, there were two datasets: a training dataset and a testing dataset. The first set of data is developed a forecasting model, and it contains 240 monthly price data from 2000 to 2019, which uses historical monthly pricing data ranging from 3 to 20 years. The second set of data has tested the model with actual price data to develop a forecast model and validate the model with current price data, consisting of 12 monthly price data in 2020. The Box-Jenkins forecasting technique was found to be suitable for forecasting the prices of Thai white rice 15%, Thai jasmine rice 105, sugar cane, cassava, and corn, while the Double Exponential Smoothing technique was shown to be suitable for forecasting peanut prices with MAPE ranging from 1.22% to 8.13%. The MAPE of field crops for the testing dataset varies from 2.00% to 11.94%, which is an accurate estimate.
 https://www.oae.go.th. (Accessed on 7 July 2021)
 S. Deepradit, P. Ongkunaruk and R. Pisuchpen, The study of forecasting techniques for aromatic coconut monthly prices using individual and hierarchical forecasting, Thai Journal of Operation Research, 2020, 8(2), 15-26. (in Thai)
 Ü.Ç. Büyükşahin and Ş. Ertekin, Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition, Neurocomputing, 2019, 361, 151-163.
 M. Ohyver and H. Pudjihastuti, Arima model for forecasting the price of medium quality rice to anticipate price fluctuations, Procedia Computer Science, 2018, 135, 707-711.
 M. Tanyarattanasrisakul, The accuracy comparison of time series model between Winters’ exponential smoothing and Box - Jenkins methods: A case study of forecasting garden coconut price, RMUTSB Academic Journal, 2018, 6(2), 101-113. (in Thai)
 W. Anggraeni, F. Mahananto, A.Q. Sari, Z. Zaini, K.B. Andri and Sumaryanto, Forecasting the price of Indonesia’s rice using hybrid artificial neural network and autoregressive integrated moving average (Hybrid NNs-ARIMAX) with exogenous variables, Procedia Computer Science, 2019, 161, 677-686.
 P. Saelim, V. Kanjanavajee, P. Suwannasean and N. Sopipan, Forecasting jasmine rice yield in Nakhon Ratchasima, Science and Technology Research Journal Nakhon Ratchasima Rajabhat University, 2019, 4(2), 25-37. (in Thai)
 N. Luangtong and N. Kantanantha, Selection of the appropriate agricultural yield forecasting models, Thai Science and Technology Journal, 2016, 24(3), 370-381. (in Thai)
 S. Kodsueb and K. Boonlha, Construction of model for the price of Thai jasmine rice 105, Science and Technology Nakhon Sawan Rajabhat University Journal, 2016, 8(8), 49-60. (in Thai)
 P. Jinno, Forecasting Thai rice export price using ARIMAX model. Thesis, Chiangmai University, Thailand, 2016.
 J. Montaño, A. Palmer, A. Sesé and B. Cajal, Using the R-MAPE index as a resistant measure of forecast accuracy, Psicothema, 2013, 25, 500-506.
 P. Ongkunaruk, Introduction to supply chain management for Agro-industry, The one printing Inc., Bangkok, Thailand, 2019.
 https:// www.ricethailand.go.th/. (Accessed on 7 July 2021)
 https://tattawin.com/ (Accessed on 7 July 2021)
 The Agricultural Research Development Agency (Public Organization), Maize, soybeans, green beans and peanuts, Thai Economic Crop Direction in Asean, Pronthip Inc., Bangkok, Thailand, 2016.
 P. Khamchoo, P. Malawal and A. Wongchai, Technical efficiency of maize production in Wiang Sa District, Nan Province, Khon Kaen Agricultural Journal, 2020, 48(1), 735-742. (in Thai)
 https://www.doa.go.th/. (Accessed on 8 July 2021)