
94 BUSINESS APPLICATIONS OF MULTIPLE REGRESSION
At this point, there is no need to compute SSR because the formula
SST
SSR + SSE = SST
allows us to compute it based on SST and SSE.
8
Nevertheless, SSR is rep-
resented by the following equation:
SSR
For row 1, that gives us 31.0 – 12.151 = 18.8 and 18.8
2
= 354.0. For row
2, 11.3 – 12.151 = – 0.9 and – 0.9
2
= 0.8. These calculations are carried
out for all the data, and the total of these squared values is 5,360.3338.
This is the SSE value shown back in Figure 3.12.
How Good Is the Fit?
In the last chapter on simple regression, we saw that r
2
represents the
percentage of the variation in the dependent ...