Forecasting Agricultural Production Using ARIMAX model

Abstract

This study explores the dynamic relationship between rainfall and agricultural production in Tamil Nadu, with a specific focus on maize, a staple crop critical to the region’s food security and economy. Employing time series analysis, the research utilizes both the Autoregressive Integrated Moving Average (ARIMA) and ARIMA with exogenous variables (ARIMAX) models to understand and forecast maize production trends. Initially, an ARIMA (p,d,q) model is developed based on historical maize production data to identify and forecast future production trends. In parallel, an ARIMAX (p,d,q) model is constructed by integrating rainfall data as an exogenous variable, enabling the study to assess the impact of climatic factors particularly precipitation on maize productivity. This dual-model approach facilitates a comparative analysis of forecast accuracy. The study not only helps in understanding the underlying patterns in maize production but also emphasizes the significance of incorporating climatic variables into predictive modelling for more realistic forecasts. Such insights are useful for policymakers, agricultural planners, and farmers which helps them make decisions on crop planning, irrigation strategies, and risk management. Ultimately, this research contributes to the development of climate-resilient agricultural strategies in Tamil Nadu, where rainfall anomalies increasingly pose a challenge to stable food production.