主成分中的综合得分是用来干嘛的为什么要算这个请简单举例分析
发布网友
发布时间：2023-04-30 00:53
共2个回答
热心网友
时间：2023-10-05 15:03
0. 这是原始数据,以备检验
90342.00 52455.00 101091.00 19272.00 82.00 16.10 197435.00 0.17
4903.00 1973.00 2035.00 10313.00 34.20 7.10 592077.00 0.00
6735.00 21139.00 3767.00 1780.00 36.10 8.20 726396.00 0.00
49454.00 36241.00 81557.00 22504.00 98.10 25.90 348226.00 0.98
139190.00 203505.00 215898.00 10609.00 93.20 12.60 139572.00 0.63
12215.00 16219.00 10351.00 6382.00 62.50 8.70 145818.00 0.07
2372.00 6572.00 8103.00 12329.00 184.40 22.20 20921.00 0.15
11062.00 23078.00 54935.00 23804.00 370.40 41.00 65486.00 0.26
17111.00 23907.00 52108.00 21796.00 221.50 21.50 63806.00 0.28
1206.00 3930.00 6126.00 15586.00 330.40 29.50 1840.00 0.44
2150.00 5704.00 6200.00 10870.00 184.20 12.00 8913.00 0.27
5251.00 6155.00 10383.00 16875.00 146.40 27.50 78796.00 0.15
14341.00 13203.00 19396.00 14691.00 94.60 17.80 6354.00 1.57
1. 这是原始数据的标准化结果
1.501 0.38474 0.9344 0.7508 -0.62952 -0.31463 0.056399 -0.47047
-0.53684 -0.55797 -0.68672 -0.6214 -1.0781 -1.2169 1.7483 -0.8466
-0.49314 -0.20006 -0.65838 -1.9283 -1.0603 -1.1066 2.3242 -0.8466
0.52579 0.081954 0.61471 1.2458 -0.47842 0.66782 0.70288 1.339
2.6662 3.2055 2.8133 -0.57606 -0.52441 -0.66551 -0.19168 0.54443
-0.36243 -0.29194 -0.55063 -1.2235 -0.81253 -1.0565 -0.1649 -0.70638
-0.59721 -0.47209 -0.58742 -0.31262 0.33151 0.2969 -0.70037 -0.51498
-0.38993 -0.16385 0.17902 1.4449 2.0771 2.1816 -0.5093 -0.26793
-0.24565 -0.14837 0.13276 1.1374 0.6797 0.22672 -0.51651 -0.239
-0.62502 -0.52143 -0.61977 0.18623 1.7017 1.0287 -0.78217 0.11933
-0.6025 -0.4883 -0.61856 -0.53608 0.32963 -0.72566 -0.75185 -0.24345
-0.52854 -0.47988 -0.5501 0.38366 -0.025123 0.82822 -0.45224 -0.5172
-0.31172 -0.34826 -0.4026 0.049154 -0.51127 -0.14421 -0.76282 2.6499
2. 这是原始数据的相关系数矩阵
1 0.91962 0.96201 0.10887 -0.28858 -0.16632 0.0067192 0.21396
0.91962 1 0.94676 -0.055032 -0.19728 -0.17094 -0.014926 0.18553
0.96201 0.94676 1 0.23295 -0.10361 0.0041839 -0.078094 0.24666
0.10887 -0.055032 0.23295 1 0.55986 0.78087 -0.44968 0.30089
-0.28858 -0.19728 -0.10361 0.55986 1 0.82664 -0.60877 -0.029523
-0.16632 -0.17094 0.0041839 0.78087 0.82664 1 -0.49215 0.17422
0.0067192 -0.014926 -0.078094 -0.44968 -0.60877 -0.49215 1 -0.29986
0.21396 0.18553 0.24666 0.30089 -0.029523 0.17422 -0.29986 1
3. 这是原始数据的特征值 (降序排列):
3.1049 2.8974 0.93022 0.64212 0.30408 0.086598 0.032184 0.0024418
4. 这是原始数据的特征向量,每列为对应于上面相应特征值的向量:
0.47665 0.29599 0.10419 0.045303 0.18422 0.065854 0.75762 0.245
0.47281 0.27789 0.16298 -0.17443 -0.30545 0.048451 -0.51841 0.52711
0.42385 0.37795 0.15626 0.05867 -0.017475 -0.099048 -0.17404 -0.78054
-0.21289 0.45141 -0.0085443 0.51609 0.53941 -0.28786 -0.24943 0.22013
-0.38846 0.33094 0.32113 -0.19942 -0.4499 -0.58229 0.23297 0.030623
-0.35243 0.40274 0.14514 0.27926 -0.31684 0.71357 0.056436 -0.042355
0.21483 -0.37741 0.14046 0.75817 -0.4182 -0.19359 0.052842 0.04116
0.055034 0.27274 -0.89116 0.071855 -0.3222 -0.12217 0.067111 -0.0032996
5. 这是判别结果,依次为: 特征值, 累计百分率, 主成分表达式
Lamda( 1)= 3.1049; PerCent = 38.81%; Y( 1) = 0.4767 * X1 + 0.4728 * X2 + 0.4238 * X3 - 0.2129 * X4 - 0.3885 * X5 - 0.3524 * X6 + 0.2148 * X7 + 0.0550 * X8
Lamda( 2)= 2.8974; PerCent = 75.03%; Y( 2) = 0.2960 * X1 + 0.2779 * X2 + 0.3780 * X3 + 0.4514 * X4 + 0.3309 * X5 + 0.4027 * X6 - 0.3774 * X7 + 0.2727 * X8
Lamda( 3)= 0.9302; PerCent = 86.66%; Y( 3) = 0.1042 * X1 + 0.1630 * X2 + 0.1563 * X3 - 0.0085 * X4 + 0.3211 * X5 + 0.1451 * X6 + 0.1405 * X7 - 0.8912 * X8
Lamda( 4)= 0.6421; PerCent = 94.68%; Y( 4) = 0.0453 * X1 - 0.1744 * X2 + 0.0587 * X3 + 0.5161 * X4 - 0.1994 * X5 + 0.2793 * X6 + 0.7582 * X7 + 0.0719 * X8
Lamda( 5)= 0.3041; PerCent = 98.48%; Y( 5) = 0.1842 * X1 - 0.3054 * X2 - 0.0175 * X3 + 0.5394 * X4 - 0.4499 * X5 - 0.3168 * X6 - 0.4182 * X7 - 0.3222 * X8
Lamda( 6)= 0.0866; PerCent = 99.57%; Y( 6) = 0.0659 * X1 + 0.0485 * X2 - 0.0990 * X3 - 0.2879 * X4 - 0.5823 * X5 + 0.7136 * X6 - 0.1936 * X7 - 0.1222 * X8
Lamda( 7)= 0.0322; PerCent = 99.97%; Y( 7) = 0.7576 * X1 - 0.5184 * X2 - 0.1740 * X3 - 0.2494 * X4 + 0.2330 * X5 + 0.0564 * X6 + 0.0528 * X7 + 0.0671 * X8
Lamda( 8)= 0.0024; PerCent = 100.00%; Y( 8) = 0.2450 * X1 + 0.5271 * X2 - 0.7805 * X3 + 0.2201 * X4 + 0.0306 * X5 - 0.0424 * X6 + 0.0412 * X7 - 0.0033 * X8
6. 这是各主成分向量、每个样本的主成分综合计算得分、排序
SAMPLE PCA1 PCA2 PCA3 PCA4 PCA5 PCA6 PCA7 PCA8 SCORE SN
SAMPLE1 1.4752 0.7586 0.5380 0.4898 1.0586 -0.0026 0.3949 0.0044 0.9910 2
SAMPLE2 0.4982 -2.5916 0.2283 0.8519 0.1606 -0.2911 -0.1272 0.0669 -0.6479 11
SAMPLE3 1.0564 -3.2255 0.4094 0.5825 -0.9300 0.0594 0.0822 -0.0240 -0.6982 13
SAMPLE4 0.4599 1.1836 -0.9977 1.5996 0.0114 0.0746 -0.0086 -0.0520 0.6207 3
SAMPLE5 4.5285 2.2624 0.4676 -0.7581 -0.4963 0.0191 -0.1211 0.0226 2.5514 1
SAMPLE6 0.3300 -1.7736 0.0311 -0.9380 0.3689 0.2062 -0.0273 -0.0668 -0.5698 10
SAMPLE7 -1.1025 -0.3179 0.2818 -0.6917 0.0914 0.3033 -0.0051 -0.0350 -0.5591 9
SAMPLE8 -2.1950 2.2441 1.0992 0.5568 -0.5719 0.0113 -0.0399 -0.0524 0.1116 4
SAMPLE9 -0.8412 0.8957 0.3529 0.1285 0.5266 -0.4687 -0.2882 -0.0009 0.0631 5
SAMPLE10 -2.0319 0.8252 0.2311 -0.5141 -0.6475 -0.1786 0.2794 0.0727 -0.5295 8
SAMPLE11 -0.7133 -0.7556 -0.1226 -1.1110 0.2343 -0.3822 0.0178 -0.0295 -0.6491 12
SAMPLE12 -1.2014 0.0303 0.2870 0.0817 0.3704 0.6423 -0.1693 0.0786 -0.3950 7
SAMPLE13 -0.2630 0.4643 -2.8063 -0.2779 -0.1766 0.0071 0.0125 0.0154 -0.2891 6
根据排序得分，可以进行判断重要性啊或者主要问题所在啊。
热心网友
时间：2023-10-05 15:04
综合得分：主要利用成分得分和方差解释率这两项指标，计算得到综合得分，用于综合竞争力对比（综合得分值越高意味着竞争力越强）。
使用在线spssau分析，可直接保存综合得分，不用计算。
综合得分
排名按照综合得分的大小进行比较，数值越大排名越高。
具体案例请见：主成分分析-SPSSAU
