In: Math
We performed a linear regression analysis between number of times on phone per drive and number of near accidents. The equation is Y= 0.320 + 0.943X, where Y is the number of times on phone per drive and X is the number of near accidents. calculate the p-value and give a conclusion.
number of times on phone per dr | near accidents |
0 | 0 |
0 | 1 |
1 | 1 |
2 | 1 |
3 | 1 |
1 | 2 |
2 | 2 |
3 | 2 |
4 | 2 |
2 | 3 |
3 | 3 |
4 | 3 |
2 | 4 |
3 | 4 |
4 | 4 |
5 | 4 |
6 | 4 |
3 | 5 |
4 | 5 |
5 | 5 |
6 | 5 |
4 | 6 |
5 | 6 |
7 | 6 |
8 | 7 |
9 | 7 |
Let's calculate the F-statistic first whose formula is:
n = 26
The calculations are given below:
y |
x | yp (Predicted) = 0.320 + 0.943*x | (y-u)*(y-u) | (yp-u)*(yp-u) | (y - yp)*(y-yp) | |
0 | 0 | 0.32 | 13.633 | 11.372 | 0.102 | |
0 | 1 | 1.263 | 13.633 | 5.902 | 1.595 | |
1 | 1 | 1.263 | 7.249 | 5.902 | 0.069 | |
2 | 1 | 1.263 | 2.864 | 5.902 | 0.543 | |
3 | 1 | 1.263 | 0.479 | 5.902 | 3.017 | |
1 | 2 | 2.206 | 7.249 | 2.209 | 1.454 | |
2 | 2 | 2.206 | 2.864 | 2.209 | 0.042 | |
3 | 2 | 2.206 | 0.479 | 2.209 | 0.630 | |
4 | 2 | 2.206 | 0.095 | 2.209 | 3.218 | |
2 | 3 | 3.149 | 2.864 | 0.295 | 1.320 | |
3 | 3 | 3.149 | 0.479 | 0.295 | 0.022 | |
4 | 3 | 3.149 | 0.095 | 0.295 | 0.724 | |
2 | 4 | 4.092 | 2.864 | 0.160 | 4.376 | |
3 | 4 | 4.092 | 0.479 | 0.160 | 1.192 | |
4 | 4 | 4.092 | 0.095 | 0.160 | 0.008 | |
5 | 4 | 4.092 | 1.710 | 0.160 | 0.824 | |
6 | 4 | 4.092 | 5.325 | 0.160 | 3.640 | |
3 | 5 | 5.035 | 0.479 | 1.803 | 4.141 | |
4 | 5 | 5.035 | 0.095 | 1.803 | 1.071 | |
5 | 5 | 5.035 | 1.710 | 1.803 | 0.001 | |
6 | 5 | 5.035 | 5.325 | 1.803 | 0.931 | |
4 | 6 | 5.978 | 0.095 | 5.224 | 3.912 | |
5 | 6 | 5.978 | 1.710 | 5.224 | 0.956 | |
7 | 6 | 5.978 | 10.941 | 5.224 | 1.044 | |
8 | 7 | 6.921 | 18.556 | 10.424 | 1.164 | |
9 | 7 | 6.921 | 28.172 | 10.424 | 4.322 | |
Average, u | 3.692308 | Total | 129.538 | 89.233 | 40.327 |
SSR = 89.233
SSE = 40.327
F = (89.233/1) / (40.327/24) = 53.09368
p-value = 1.58E-07 (at df = 1, 24)
As the p-value < 0.05, we can conclude that the regression is significant at 0.01 level of significance.
SSE =