Times Series Analysis of Variance to Investigate the Effect of Rainfall Patterns on Rice and Maize Production in Nasarawa State

Abstract

Agriculture in Nasarawa state is predominantly rain-fed, making crop output highly sensitive to climatic variability. This study was conducted to examine the effect of annual rainfall patterns on rice and maize production in Nasarawa State, Nigeria. Secondary data sourced from the Nasarawa Agricultural Development Programme (NADP) from 2014 to 2023 was used. The research employs time series regression analysis to quantify the statistical relationship between annual rainfall and crop yields. Descriptive statistics and Augmented Dickey Fuller tests showed that rainfall, rice, and maize series were non-stationary; however, ordinary least squares (OLS) models were estimated with appropriate diagnostic checks. The regression results indicate a positive and statistically significant relationship between rainfall and the yields of both rice (p = 0.0116) and maize (p = 0.0076). Rainfall accounted for approximately 57% and 61% of the variation in rice and maize production, respectively. Diagnostic tests show normally distributed residuals, though mild autocorrelation and non-stationarity suggest caution in causal interpretation. The findings confirm that rainfall is a key climatic factor influencing staple crop production in Nasarawa State. The use of more advanced time series techniques such as cointegration analysis, ARIMA/ARIMAX, or VAR models and inclusion of additional agricultural variables to strengthen predictive accuracy was recommended. Overall, the research provides evidence-based insights to support agricultural planning, climate adaptation strategies, and policy formulation in the state.