Agricultural Forecasting in a Changing Climate: ARIMA-X Model of Cereal Production in Tanzania

This study investigates the forecasting of cereal production in Tanzania using an Autoregressive Integrated Moving Average (ARIMA) model integrated with exogenous variables (ARIMAX). Annual time series data spanning from 1960 to 2022 were obtained from the World Bank's World Development Indicators (WDI) and include cereal production, land under cereal and arable cultivation, working-age population, average annual precipitation, and temperature. To ensure robustness in modeling, missing values were imputed through linear interpolation, and stationarity was verified using the Augmented Dickey-Fuller (ADF) test. First and second-order differencing was applied where necessary to achieve stationarity.An ARIMA model with exogenous regressors differenced land used, arable land, and temperature was fitted after excluding statistically insignificant predictors to reduce multicollinearity. Forecasts from 2023 to 2027 predict fluctuating trends in cereal production, ranging from 10.2 to 11.38 million metric tons. The positive and significant influence of temperature highlights the role of climatic factors in cereal production, while land use variables show moderate effects. The findings underscore that Tanzania’s cereal production is shaped by structural trends, calling for dynamic, climate-responsive policy frameworks. The study provides a data-driven foundation for policymakers to develop adaptive strategies ensuring food security and economic stability amidst climatic and demographic shifts.