<![CDATA[Coconut is a strategic plantation commodity in Padang Pariaman Regency, Indonesia, whose productivity is increasingly influenced by climate variability. Accurate, climate-responsive forecasting is therefore essential to support production planning and early warning systems. This study aims to develop and evaluate a climate-informed ARIMAX model for forecasting coconut production and assessing its potential application in an early warning framework. Annual coconut production data for 2014–2024 were combined with climate variables, including rainfall, temperature, humidity, wind speed, and the number of rainy days. A baseline ARIMA model was first identified, followed by ARIMAX modeling using Maximum Likelihood Estimation. Model selection was based on Akaike Information Criterion (AIC), while forecasting performance was evaluated using Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). The results showed that ARIMA(1,1,1) was identified as the optimal baseline model and achieved the lowest forecasting errors based on RMSE and MAPE, indicating its strong capability in capturing the temporal pattern of coconut production. The incorporation of climate variables through ARIMAX demonstrated that rainfall and the number of rainy days significantly influenced coconut production, while temperature, humidity, and wind speed exhibited weaker effects. To improve model stability and avoid multicollinearity, a simplified log-ARIMAX(1,1,1) model was developed by retaining rainfall as the primary exogenous variable. This model achieved the lowest AIC value among the evaluated climate-based models, indicating improved parsimony and explanatory capability. Forecasting results for 2025–2029 indicate a moderate and continuous increase in coconut production. Although ARIMA provides superior predictive accuracy, the rainfall-based ARIMAX model offers additional insights into climate–production relationships, making it valuable for supporting climate-informed forecasting and the future development of early warning frameworks for coconut plantation productivity.]]>
