Sample | X | Y |
1 | 5.4 | 7537 |
2 | 5.7 | 5440 |
3 | 1.2 | 4322 |
4 | 0.1 | 1681 |
5 | 4.2 | 5612 |
6 | 4.0 | 5756 |
7 | 4.5 | 7209 |
8 | 9.0 | 6064 |
9 | 2.2 | 5101 |
10 | 5.7 | 7799 |
11 | 3.9 | 6491 |
12 | 8.4 | 6252 |
13 | 4.4 | 6756 |
14 | 8.3 | 8480 |
15 | 5.0 | 5834 |
16 | 1.4 | 4217 |
17 | 9.6 | 8757 |
18 | 9.2 | 8254 |
19 | 6.7 | 6448 |
20 | 2.5 | 3843 |
- Create an appropriately formatted and labeled scatter diagram of these data. Discuss what you observe about the pattern of the data and discuss what one should look for to decide whether a simple linear regression model is appropriate.
- Assuming that a simple linear regression model is appropriate, fit a least squares regression model. Show the detailed equations and partial sums in the equations for the regression coefficients.
- Test for significance of regression at a = 0.05. (It is your choice what technique you want to use.)
- Plot the residuals versus and comment on the underlying regression assumptions. Specifically, does it appear that the equality of variance assumption is satisfied?
- Plot the residuals on a normal probability plot and comment on the underlying regression assumptions. Specifically, does it appear that the residuals are approximately normal?
- If a measurement of X has a value of 4.4, determine what the predicted value of Y is and explain what that value represents.
- Estimate how likely it is that an observed value of Y greater than 6756 occurs when X = 4.4.
No comments:
Post a Comment