<![CDATA[Monthly inflation in North Sumatra exhibits high fluctuation, seasonality, and uncertainty (stochasticity), making it difficult for deterministic models to predict accurately. This study aims to analyze seasonal and stochastic patterns, build a SARIMA model, and apply Monte Carlo Simulation to generate probabilistic prediction ranges. This study uses a quantitative approach with BPS monthly inflation (month-to-month) data from January 2021–September 2025 for five cities in North Sumatra. The Seasonal Autoregressive Integrated Moving Average (SARIMA) model is used to capture seasonal patterns and trends, while Monte Carlo Simulation (5,000 paths) is applied to quantify prediction uncertainty by generating probabilistic distributions. The forecast results (Oct 2025–Dec 2026) indicate that Medan City is projected to face the highest inflationary pressure. However, model validation on 2024 test data showed limited performance, marked by negative R² values for all cities (ranging from -0.364 to -1.146). This negative R² finding highlights high uncertainty. This combined approach proves more informative than single-point predictions, as it explicitly presents the uncertainty range (visualized as a "fan 90%" band), offering a more comprehensive forecasting picture for policy considerations.]]>
Copyrights © 2025
