H0: β2, β2, β2 =0
H1= β2, β2, β2 ≠ 0
F= 2.45 in Excel, significance threshold = 0.05
We reject the alternative hypothesis since F (2.45)>significant level (0.05) and conclude that there is a relationship between auto rate and age, male percentage, and female %.
R2
The excel table results show that R2 = 0.1378, indicating that age, percent males, and percent males explain 13.78% of the variation in auto rates. Because the R2 squared is tiny, we conclude that there are only weak relationships between auto rates and age, males, and females. The marginal effect indicates that there is are a relationship between the variables. From excel, a unit change in male percentage results in a negative change in auto rates by one 35017.74. Similarly, a change in age by one unit causes the auto rates to change by 33.54 in a positive direction. Finally, a unit change of female percentage results in 3.84 change in auto rates.
T-statistics
Hypothesis statement
H0: β2, β2, β2 =0
H1= β2, β2, β2 ≠ 0
Significant level = 0.05
The result of the excel shows that age has a p-value of 0.470 which is higher than significant level of 0.05. We conclude that there is no significant relationship between auto rates and age. Similarly, percent female has a p-value of 0.822 which is higher than the significant level of 0.05. We conclude that there is no significant relationship between auto rates and percent females. However, a p-value of percent male is 0.0228. The value is less than an alpha of 0.05(0.0228
