Abstract
This study conducts a comprehensive statistical investigation of trends and relationships between economic indicators in Suleja, Nigeria, from 2019 to 2023. Employing inferential statistics, data visualization techniques, and a robust regression model with diagnostic checks, we uncover underlying patterns and relationships. Our analysis reveals significant relationships between economic variables, identifying nonlinear relationships and highlighting the importance of accounting for multicollinearity, autocorrelation, and heteroscedasticity in economic modeling. Linear regression analysis reveals a robust model with no significant autocorrelation in the residuals (Durbin-Watson statistic = 0.213), a high R-squared value (R² = 0.999), and a low Root Mean Squared Error (RMSE = 2.5). The ANOVA table shows a significant F-statistic (F = 2976.330, p < 0.001) and a high R-squared value (R² = 0.999), indicating a significant improvement in the fit of the alternative model. Coefficient analysis reveals significant coefficients for V2023 (p = 0.008) and no multicollinearity between independent variables, with tolerance values ranging from 0.000 to 1.000 and variance inflation factor (VIF) values ranging from 1.000 to 6933.238. Descriptive statistics show increasing means (range: 12.4 to 234.5) and standard deviations (range: 2.1 to 89.4) for economic variables over time. The covariance matrix reveals positive relationships between certain variables, with covariance values ranging from 0.124 to 0.254. Collinearity diagnostics indicate potential multicollinearity issues, with condition indices ranging from 1.000 to 6933.238. Casewise diagnostics identify influential data points, with Cook's distances ranging from 0.000 to 7.512. Residual statistics show a good fit for the regression model, with a mean standardized residual of 0.098 and a standard deviation of 1.312. Our findings contribute to the existing literature on economic relationships, highlighting the importance of rigorous statistical analysis in understanding economic trends and relationships. Our approach demonstrates the effectiveness of regression analysis in modeling economic relationships, providing a framework for future research and policy analysis in Suleja, Nigeria.