Answer:
0.75
0.7205882
25.5
Result is significant at α = 0.01 and α = 0.05
Model improved
Step-by-step explanation:
Given that:
Number of observations (n) = 20
Total sum of squares (SST) = 1000
Model sum of squares (SSR) = 750
1) R² = SSR / SST = 750 / 1000 = 0.75
2.)
Adjusted R² = [(SST - SSR) /(n-k-1)] / (SST ÷ (n - 1))
k = number of regressors = 2
Adj R² = 1 - ((1000 - 750) / (20-2-1)) / (1000 / (20 - 1))
1 - 0.2794117 = 0.7205882
3.) Fstat = (SSR / k) / ((SST - SSR) / (n - k-1))
= (750 /2) / ((1000 - 750) / (20 - 2 - 1))
= 25.5
4.) At α = 0.05
Fα,k,(n - k-1) = F0.05, 2, (20 - 2 - 1) = F0.05,2, 17 = 3.5915 (f distribution calculator)
Fstat > F0.05, 2, (20 - 2 - 1)
25.5 > 3.5915 (Hence result is significant at α = 0.05
At α = 0.01
Fα,k,(n - k-1) = F0.01, 2, (20 - 2 - 1) = F0.01,2, 17 = 6.112 (f distribution calculator)
Fstat > F0.01, 2, (20 - 2 - 1)
25.5 > 6.112 (Hence result is significant at α = 0.01
Adjusted R² if a 3rd regressors is added : k = 3
Adjusted R² = [(SST - SSR) /(n-k-1)] / (SST ÷ (n - 1))
k = number of regressors = 3
SSR = 785
Adj R² = 1 - ((1000 - 785) / (20-3-1)) / (1000 / (20 - 1))
1 - 0.2553125 = 0.7446875
Adjusted R² value is now 0.7446875 which is greater than with 2 regressors,. Hence, adding a third regressors improved the model.
