Forecasting the price of Indonesia's rice using hybrid artificial neural network and autoregressive integrated moving average (hybrid NNS-ARIMAX) with exogenous variables

As a primary food, rice has a special attention by the Indonesian Government. The variability and trend of rice price become its main concern. Based on the data obtained from Indonesian national statistics agency, it shows that there is an increasing trend toward the retail price of rice in traditional markets. The price of rice has uniqueness in the process of determining it. Many variables have influenced the price and it is highly regulated. In order to help the decision maker to determine the price, they somehow need a clear insight of future trend of its price changing regarding to several influencing variable. Thus, an appropriate forecasting should be conducted. This research includes rice harvest area, rice production, rice consumption, season as independent variables and use combination of Artificial Neural Network and ARIMAX to forecast the price of rice in in several Indonesian provinces. The result shows that the combination model gives better result than ANN model. The average of decreasing MAPE about 1.21% for ANN and Hybrid NNs-ARIMA, and 0.23% for ANN and Hybrid NNs-ARIMAX. The results of this research are expected to help the Ministry of Agriculture and the National Logistics Agency in making decisions and policies of national rice price.