放送大学「心理統計法」第14章 多変量解析のすすめ
|
|
固有値
|
|
Comp.1 Comp.2 Comp.3 Comp.4 Comp.5 Comp.6 Comp.7 Comp.8
3.3564371 1.3426892 1.0813617 0.8056899 0.5029603 0.4290183 0.3509757 0.1308679
説明率、累積説明率
|
|
Importance of components:
Comp.1 Comp.2 Comp.3 Comp.4 Comp.5 Comp.6 Comp.7 Comp.8
Standard deviation 1.8320581 1.1587447 1.0398854 0.8976023 0.70919692 0.65499492 0.59243201 0.36175662
Proportion of Variance 0.4195546 0.1678361 0.1351702 0.1007112 0.06287003 0.05362729 0.04387196 0.01635848
Cumulative Proportion 0.4195546 0.5873908 0.7225610 0.8232722 0.88614226 0.93976956 0.98364152 1.00000000
主成分負荷量 (テキストとは符号が逆の可能性あり)
|
|
Comp.1 | Comp.2 | Comp.3 | |
---|---|---|---|
みずみずしさ | 0.34362 | 0.27407 | 0.68568 |
なめらかさ | -0.77847 | 0.25826 | 0.09472 |
しみこみの速さ | 0.08675 | 0.89300 | -0.10309 |
さらさら感 | 0.76950 | -0.15316 | 0.07733 |
しっとり感 | -0.83347 | 0.01943 | -0.30085 |
つるつる感 | -0.54494 | 0.20069 | 0.59096 |
やわらかさ | -0.88216 | 0.02886 | -0.05928 |
香りの好み | -0.51270 | -0.58179 | 0.37732 |
主成分得点(標準化したもの_テキストとは符号が逆の可能性あり)
|
|
Comp.1 | Comp.2 | Comp.3 | |
---|---|---|---|
1 | -1.5461 | -1.8890 | -0.6410 |
2 | -0.8453 | -0.3072 | 1.5418 |
3 | 0.0834 | 1.0938 | 0.7851 |
4 | 2.3658 | -1.8230 | -0.7070 |
5 | -0.0855 | 1.3763 | 1.2171 |
6 | 1.7742 | 0.6595 | 0.1993 |
7 | -0.3864 | 1.1818 | -1.1294 |
8 | 1.0160 | 1.1611 | 0.0673 |
9 | -0.2849 | -0.1279 | 0.5954 |
10 | -0.7638 | -0.5998 | -0.2096 |
11 | -1.3976 | 1.6809 | -0.1309 |
12 | 0.3857 | 1.0092 | 0.4280 |
13 | 1.2944 | 0.2204 | 2.6219 |
14 | -0.0390 | 0.1999 | -1.6683 |
15 | -0.6255 | 0.9173 | -1.0699 |
16 | 1.1216 | -0.3757 | -1.6511 |
17 | 0.7568 | 0.9921 | -0.3891 |
18 | -0.0126 | 0.2593 | -1.6781 |
19 | -0.5316 | -0.4739 | 0.7746 |
20 | -0.4341 | -1.2468 | 0.8008 |
21 | -0.3225 | -1.6447 | 0.8293 |
22 | -1.4197 | 0.4786 | 0.3854 |
23 | -0.7684 | -0.8773 | -0.3732 |
24 | -0.2174 | -0.8412 | -0.3134 |
25 | -0.5736 | 0.5708 | -0.8040 |
26 | -0.0069 | -0.5150 | 0.2585 |
27 | -0.4222 | -0.6757 | 0.3935 |
28 | 1.8850 | -0.4037 | -0.1329 |
主成分得点(標準化したもの)上位3つをデータセットdに付け加える
|
|
AICを用いて変数選択
|
|
変数選択
Start: AIC=-4.37
心地よさ ~ PC1 + PC2 + PC3
Df Sum of Sq RSS AIC
- PC3 1 0.0009 18.002 -6.3689
18.001 -4.3703 - PC2 1 3.6362 21.637 -1.2186
- PC1 1 24.4694 42.470 17.6647
Step: AIC=-6.37
心地よさ ~ PC1 + PC2
Df Sum of Sq RSS AIC
- PC2 1 3.6362 21.638 -3.2175
- PC1 1 24.4694 42.471 15.6653
結果
Estimate | Std. Error | t value | Pr(>|t|) | |
---|---|---|---|---|
(Intercept) | 3.8214 | 0.1604 | 23.83 | 0.0000 |
PC1 | -0.9520 | 0.1633 | -5.83 | 0.0000 |
PC2 | -0.3670 | 0.1633 | -2.25 | 0.0337 |