Predictive Time-Series Modelling of Rice Price Fluctuations in East Nusa Tenggara Using ARIMAX: A Data Driven Case Study Regional Agricultural Market Dynamics
Main Article Content
Abstract
Rice prices are a crucial commodity affecting the economy and the well-being of the community, making accurate forecasting essential to support effective policies. In the Province of East Nusa Tenggara (NTT), rice prices are notably high, reaching IDR 14,000 per kilogram in the city of Kupang and up to IDR 17,000 per kilogram in rural areas. This increase is due to high demand, crop failures in several regions, and dwindling stock availability. This study aims to forecast rice prices in Kupang, Maumere, and Sumba Timur cities in the Province of East Nusa Tenggara (NTT) using the ARIMAX method. Data on rice prices from 2018 to 2024 were used in this study, with exogenous variables including religious holidays. The ARIMAX method was chosen for its ability to incorporate exogenous variables affecting rice prices. The study projects rice prices in NTT for the next two years using the ARIMAX model with auto_arima parameters (2, 1, 3). The ARIMAX model, an extension of ARIMA with added exogenous variables, has proven effective in providing accurate predictions. The results show that the ARIMAX model generates highly accurate rice price forecasts, with a Mean Absolute Percentage Error (MAPE) of 0.01%.