It can be very misleading to rely on the correlation coefficient alone when selecting a regression model. To illustrate, (a) run a linear regression on the data set given (without doing a scatterplot), and note the strength of the correlation (the correlation coefficient). (b) Now run a quadratic regression (CALC 5:QuadReg) and note the strength of the correlation. (c) What do you notice? What factors other than the correlation coefficient must be taken into account when choosing a form of regression?
x | y |
50 | 67 |
100 | 125 |
150 | 145 |
200 | 275 |
250 | 370 |
300 | 550 |
350 | 600 |
We need at least 10 more requests to produce the solution.
0 / 10 have requested this problem solution
The more requests, the faster the answer.