Methodological Evaluation of Smallholder Farms Systems in Uganda Using Time-Series Forecasting Models for Reliability Assessment

Smallholder farms in Uganda face challenges such as unpredictable weather patterns, which affect their productivity and sustainability. A comprehensive search strategy was employed to identify relevant studies, focusing on methodologies such as ARIMA (AutoRegressive Integrated Moving Average) and SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous regressors). The analysis revealed a significant proportion of model accuracy improvements when using SARIMAX over traditional ARIMA models in forecasting smallholder farm yields. SARIMAX outperformed ARIMA, demonstrating higher reliability and precision for forecasting yield variability under different climatic conditions. Future research should consider integrating exogenous variables to enhance the predictive power of time-series models for smallholder farms in Uganda. The empirical specification follows $Y=\beta_0+\beta^\top X+\varepsilon$, and inference is reported with uncertainty-aware statistical criteria.