In: Statistics and Probability
a. Estimate the regression equation and determine the predicted values of y
b. Use the predicted values and the actual values of y to calculate the residuals
c. Plot the residuals (vertical axis) against the predicted values (horizontal axis).
d. Does it appear that heteroscedasticity is a problem? Explain
e. Draw a histogram of the residuals. Does it appear that the errors are normally distributed? Explain.
f. Use the residuals to identify possible outlier(s)
Weight | Price |
27 | 12.02 |
28.5 | 12.04 |
30.8 | 12.32 |
31.3 | 12.27 |
31.9 | 12.49 |
34.5 | 12.7 |
34.7 | 13 |
37 | 13 |
41 | 13.17 |
41 | 13.19 |
38.8 | 13.22 |
39.3 | 13.27 |
34 | 12.8 |
please do not use minitab or any statistical software
a. Estimate the regression equation and determine the predicted values of y:
We are given:
Obs. | WeightWeight | PricePrice | X_i^2Xi2 | Y_i^2Yi2 | X_i \cdot Y_iXi⋅Yi |
1 | 27 | 12.02 | 729 | 144.4804 | 324.54 |
2 | 28.5 | 12.04 | 812.25 | 144.9616 | 343.14 |
3 | 30.8 | 12.32 | 948.64 | 151.7824 | 379.456 |
4 | 31.3 | 12.27 | 979.69 | 150.5529 | 384.051 |
5 | 31.9 | 12.49 | 1017.61 | 156.0001 | 398.431 |
6 | 34.5 | 12.7 | 1190.25 | 161.29 | 438.15 |
7 | 34.7 | 13 | 1204.09 | 169 | 451.1 |
8 | 37 | 13 | 1369 | 169 | 481 |
9 | 41 | 13.17 | 1681 | 173.4489 | 539.97 |
10 | 41 | 13.19 | 1681 | 173.9761 | 540.79 |
11 | 38.8 | 13.22 | 1505.44 | 174.7684 | 512.936 |
12 | 39.3 | 13.27 | 1544.49 | 176.0929 | 521.511 |
13 | 34 | 12.8 | 1156 | 163.84 | 435.2 |
Sum = | 449.8 | 165.49 | 15818.46 | 2109.1937 | 5750.275 |
=1/n∑Xi=449.8/13=34.6
=34.6
=1/n∑Yi=165.49/13=12.73
=12.73
SSXX=∑Xi2−1/n(∑Xi)2=15818.46−1/13(449.8)2=255.38
SSXX=255.38
SSYY=∑Yi2−1/n(∑Yi)2=2109.1937−1/13(165.49)2=2.506
SSYY=2.506
SSXY=∑XiYi−1/n(∑Xi)(∑Yi)=5750.275−1/13(449.8)(165.49)=24.321
SSXY=24.321
=SSXY/SSXX=24.321/255.38=0.0952
=0.0952
==12.73−0.0952×34.6=9.4349
=9.4349
the regression equation is:
PREDICTED VALUES:
Obs. | Weight | Price | Predicted Values |
1 | 27 | 12.02 | 9.4349+0.0952×27=12.006 |
2 | 28.5 | 12.04 | 9.4349+0.0952×28.5=12.149 |
3 | 30.8 | 12.32 | 9.4349+0.0952×30.8=12.368 |
4 | 31.3 | 12.27 | 9.4349+0.0952×31.3=12.416 |
5 | 31.9 | 12.49 | 9.4349+0.0952×31.9=12.473 |
6 | 34.5 | 12.7 | 9.4349+0.0952×34.5=12.72 |
7 | 34.7 | 13 | 9.4349+0.0952×34.7=12.74 |
8 | 37 | 13 | 9.4349+0.0952×37=12.959 |
9 | 41 | 13.17 | 9.4349+0.0952×41=13.34 |
10 | 41 | 13.19 | 9.4349+0.0952×41=13.34 |
11 | 38.8 | 13.22 | 9.4349+0.0952×38.8=13.13 |
12 | 39.3 | 13.27 | 9.4349+0.0952×39.3=13.178 |
13 | 34 | 12.8 | 9.4349+0.0952×34=12.673 |
b. Use the predicted values and the actual values of y to calculate the residuals:
Price(observed value) | Predicted Values | Residuals=observed value-predicted value |
12.02 | 12.006 | 0.014 |
12.04 | 12.149 | -0.109 |
12.32 | 12.368 | -0.048 |
12.27 | 12.416 | -0.146 |
12.49 | 12.473 | 0.017 |
12.7 | 12.72 | -0.02 |
13 | 12.74 | 0.26 |
13 | 12.959 | 0.041 |
13.17 | 13.34 | -0.17 |
13.19 | 13.34 | -0.15 |
13.22 | 13.13 | 0.09 |
13.27 | 13.178 | 0.092 |
12.8 | 12.673 | 0.127 |
c. Plot the residuals (vertical axis) against the predicted values (horizontal axis):
d. Does it appear that heteroscedasticity is a problem? Explain:
Yes,heteroscedasticity is a problem Heteroscedasticity means unequal scatter.From the Scatter-plot we observe that the variance of the residuals is lower for lower predicted values and higher for higher predicted values.which shows a highly unequal scatter. hence, yes heteroscedasticity is a problem.
e. Draw a histogram of the residuals. Does it appear that the errors are normally distributed? Explain.
Frequency Table | |
Class | Count |
-0.2--0.071 | 4 |
-0.07-0.059 | 5 |
0.06-0.189 | 3 |
0.19-0.319 | 1 |
The normal distribution curve is symmetric that is but since the above histrogram is positively skewed hence,the errors are not normally distributed.
please rate my answer and comment for doubts.