Modeling Autoregressive Integrated Moving Average (ARIMAX) in Quarterly Agricultural GDP and Key Subsectors of Agriculture in Nigeria

Abstract

Agriculture remains a critical driver of economic growth, employment generation, and food security in Nigeria, yet the sector continues to experience structural volatility and productivity challenges. Accurate modelling and forecasting of Agricultural Gross Domestic Product (GDP) are therefore essential for effective economic planning and policy formulation. This study examines the dynamic behaviour of Nigeria’s Agricultural GDP and evaluates the contributions of its major sub-sectors - crop production, livestock, forestry, and fishing - using the Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) modelling framework. Annual time series data on Agricultural GDP and its sub-sectoral components were analyzed using modern time series econometric techniques. Preliminary statistical diagnostics, including descriptive analysis and unit root tests, were conducted to determine the stochastic properties and stationarity of the variables. Following differencing to achieve stationarity, alternative ARIMAX model specifications were estimated and evaluated using standard model selection criteria such as the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC). The empirical results identified the ARIMAX(0,1,1) model as the most parsimonious and statistically adequate specification for modelling Agricultural GDP. The findings indicate that variations in Agricultural GDP are significantly influenced by short-run shocks and by the structural contributions of the agricultural sub-sectors, with crop production emerging as the dominant driver of sectoral output. Forecast results further reveal a sustained but moderate growth trajectory for Nigeria’s Agricultural GDP within the forecast horizon, suggesting continued sectoral resilience despite macroeconomic uncertainties. The study concludes that integrating sectoral sub-components as exogenous variables improves forecasting performance and provides deeper insight into the structural dynamics of agricultural output. Consequently, the study recommends strengthened investment in crop production systems, improved livestock productivity, and enhanced data-driven agricultural policy frameworks to support sustainable sectoral growth and national economic diversification.