In: Statistics and Probability
#19: Lemons and car crashes using the listed lemon/crash data, find the best predicted crash fatality rate for a year in which there are 500 metric tons of lemon imports. Is the predication worthwhile?
Lemon Imports |
230 |
265 |
358 |
480 |
530 |
Crash Fatality rate |
15.9 |
15.7 |
15.4 |
15.3 |
14.9 |
Let x be the lemon imports in metric tons.
Let y be the crash fatality rate.
The calculation table is shown below:
Serial No. |
x |
y |
x^2 |
y^2 |
xy |
1 |
230 |
15.9 |
52900 |
252.81 |
3657 |
2 |
265 |
15.7 |
70225 |
246.49 |
4160.5 |
3 |
358 |
15.4 |
128164 |
237.16 |
5513.2 |
4 |
480 |
15.3 |
230400 |
234.09 |
7344 |
5 |
530 |
14.9 |
280900 |
222.01 |
7897 |
Total |
1863 |
77.2 |
762589 |
1192.56 |
28571.7 |
Here, n = 5.
The simple regression model is given by,
y = a + bx ----(1)
First calculate slope (b) and intercept (a) as shown below:
Now substitute the values in equation (1), we get
y = 16.4833 – 0.0028x
The best predicted crash fatality rate for a year in which there are 500 metric tons of lemon imports can be calculated as,
y = 16.4833 – 0.0028x
y = 16.4833 – 0.0028(500)
y = 16.4833 – 1.4
y=15.0833
Therefore, the best predicted crash fatality rate for a year in which there are 500 metric tons of lemon imports is 15.0833.
We will calculate the coefficient of determination (R2) to determine whether the predication is worthwhile or not as R2 considered as accuracy measure for regression model.
Since, the R2 value 0.92 which is close to 1, so it can be said that the accuracy of the above model is high.
Also, R2 value can be interpreted as that 92% of variations in crash fertility rate is explained by the lemon reports.
Therefore, it can be said that the predication is worthwhile.