Function
Ward Method 1 2 3 4
1 -2.468 .113 1.879 3.378
2 -.312 .697 .708 -2.869
3 -.255 1.835 -.549 .339
Unstandardized canonical discriminant functions evaluated at group means
上表显示出了三维的各类别重心在空间中的坐标位置。
Classification Statistics
Processed Excluded
Missing or
out-of-range group codes
At least one missing discriminating variable
Used in Output
60 0 0 60
Classification Processing Summary
Ward Method
毛利 投资 店堂面积 P15 P25 P35 P45 P55 TOTAL
1 -.368 -.103 .779 .024 -.006 -.009 -.018 -.003 .007
2 -.375 -.117 .844 .029 -.004 -.009 -.013 .000 .005
3 -.366 -.117 .861 .026 -.004 -.007 -.013 -.001 .006
16 4 -.371 -.095 .860 .032 -.005 -.010 -.011 -.001 .006
Classification Function Coefficients
平均收入 非餐饮企 月租金 生活指数 (Constant)
6.895 .886 1.554 2.282 -290.031
6.684 .863 2.196 2.329 -311.410
7.096 .869 1.968 2.192 -314.040
7.252 .845 1.675 2.203 -346.514
Fisher's linear discriminant functions
通过上表可以得到四个贝叶斯判别函数，把新样本的值对应带入四个函数式，函数值最高的即为相应分类
Predicted Group Membership
Original
Count
Ward Method 1 2 3 4
%
1 2 3 4
Cross-validated(a)
Count
1 2 3 4
%
1 2 3 4
functions derived from all cases other than that case. b 98.3% of original grouped cases correctly classified. c 81.7% of cross-validated grouped cases correctly classified.
1
23 0 0 0 95.8 .0 .0 .0 22 0 1 0 91.7 .0 4.5 .0
2
0 9 0 0 .0 100.0
.0 .0 0 8 4 0 .0 88.9 18.2 .0
3
1 0 22 0 4.2 .0 100.0
.0 2 1 16 2 8.3 11.1 72.7 40.0
4
0 0 0 5 .0 .0 .0 100.0
0 0 1 3 .0 .0 4.5 60.0
Total
24 9 22 5 100.0 100.0 100.0 100.0 24 9 22 5 100.0 100.0 100.0 100.0
Classification Results(b,c)
a Cross validation is done only for those cases in the analysis. In cross validation, each case is classified by the
根据输出结果的不一致表明利用Bayes判别方法，其判别分析的结果不稳定。
17