Monthly Rainfall Forecasting in Pesisir Selatan Regency Using the Autoregressive Integrated Moving Average (ARIMA) Model

Abstract

Rainfall is a climate variable that plays a crucial role in agricultural planning, water resource management, and hydrometeorological disaster mitigation. Therefore, a forecasting method capable of adequately describing the temporal patterns of rainfall data is required. This study aims to forecast monthly rainfall in Pesisir Selatan Regency using the Autoregressive Integrated Moving Average (ARIMA) method. The data used in this study are monthly rainfall data for the period 2015–2024. The analysis stages include missing data imputation, Box–Cox transformation, stationarity testing using the Augmented Dickey–Fuller (ADF) test, model identification through ACF and PACF plots, parameter estimation, and model evaluation based on the Akaike Information Criterion (AIC), residual diagnostic tests, and forecasting accuracy using Mean Absolute Percentage Error (MAPE). The results show that the ARIMA(0,1,1) model is the best model, as indicated by the lowest AIC value and residuals that satisfy the white noise assumption. The forecasting accuracy evaluation yields a MAPE value of 55.05%, indicating that the model’s ability to capture monthly rainfall variability is still limited. Rainfall forecasting for the period January to December 2025 produces relatively constant forecast values, reflecting the limitations of the ARIMA(0,1,1) model in representing seasonal variations. Therefore, this model is more suitable as a baseline approach for rainfall forecasting in Pesisir Selatan Regency. Future studies are recommended to apply models that incorporate seasonal components or external variables to improve forecasting accuracy.