The Impact of Sunlight, Ultraviolet Radiation, and Temperature Variability on COVID-19 Mortality Rates: Evidence from Small Countries and U.S. States and Territories

0 500 1000 1500 2000 2500 Predicted Values 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 Actual Values Actual vs Predicted (SunHoursMin) 0 500 1000 1500 2000 2500 Predicted -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 Standardized Residuals Standardized Residuals (SunHoursMin) -0.1 0 0.1 0.2 0.3 0.4 0.5 Adjusted SunHoursMin -2000 -1000 0 1000 2000 3000 4000 5000 Adjusted TotalDeath Partial Regression Plot (SunHoursMin) Adjusted data Fit: y = -4444.85*x 95% conf. bounds Durbin–Watson Statistic DW = 1.2664 Interpretation: 1 = positive autocorrelation 2 = no autocorrelation 4 = negative autocorrelation Breusch–Pagan Test LM Statistic: 3.3116 p-value: 0.0688 Conclusion: No Heteroskedasticity White Test White Statistic: 3.9695 df: 2 p-value: 0.1374 Conclusion: No Heteroskedasticity -3000 -2000 -1000 0 1000 2000 Residuals 0 0.5 1 1.5 2 2.5 3 Probability density Histogram of Residuals (SunHoursMin) -3 -2 -1 0 1 2 3 Expected N(0,1) -3 -2 -1 0 1 2 3 Observed Residuals P–P Plot (Normality Test) – SunHoursMin