认为自己是高手的进来-关于残差

1.计算Lack of Fit 和Pure Error的过程?
2.残差中 Lack of Fit 和Pure Error在回归图形中如何反映？
3.如果是多元回归又将如何呢?
这个题目还是简单了些!

The regression equation is
y1 = - 21.3 + 35.2 x1SPC

Predictor Coef SE Coef T PSPC
Constant -21.337 3.236 -6.59 0.000bbs
x1 35.248 4.125 8.55 0.000
S = 0.3540 R-Sq = 80.2% R-Sq(adj) = 79.1%

Analysis of Variance
Source DF SS MS F P
Regression 1 9.1532 9.1532 73.02 0.000
Residual Error 18 2.2563 0.1253
Lack of Fit 8 1.8413 0.2302 5.55 0.007
Pure Error 10 0.4150 0.0415
Total 19 11.4095