Time-Series Forecasting Model Evaluation for Cost-Effectiveness in Smallholder Farms Systems in Kenya,

Smallholder farms in Kenya face challenges in cost-effectiveness due to variable weather conditions and market fluctuations. A time-series analysis was conducted using autoregressive integrated moving average (ARIMA) models with robust standard errors estimated via bootstrapping techniques. The study used historical data on input costs, output prices, and weather conditions from to assess the model's predictive accuracy and cost-effectiveness. The ARIMA model demonstrated a coefficient of determination (R²) of 0. 78 for forecasting farm income over subsequent years, indicating strong explanatory power. The model was particularly effective in predicting income trends during drought periods (ranging from -15% to +30%) compared to baseline forecasts. The ARIMA model proved reliable and cost-effective for assessing the financial viability of smallholder farms under varying conditions. Future research should explore integration with climate prediction models. Farmers and policymakers should utilise this forecasting tool to make informed decisions regarding input purchases, insurance, and diversification strategies.