[回归分析求助] 如何对载荷矩阵进行旋转 [推广有奖]

11楼

gg19861026 发表于 2014-9-21 17:00:33

饭团儿发表于 2014-9-21 16:52
你好我用了rotate出来的结果也不对能否问下这是什么原因呢？

你首先需要进行因子分析，如：factor bg2cost1-bg2cost6, factors(2)，之后再用rotate

12楼

饭团儿

发表于 2014-9-21 17:17:34

gg19861026 发表于 2014-9-21 17:00
你首先需要进行因子分析，如：factor bg2cost1-bg2cost6, factors(2)，之后再用rotate

您好！我是先用pca 然后出来这样的结果

. pca liang e cang zhenfu jijinyijialv

Principal components/correlation                Number of obs =    13060
                                                Number of comp.  =       5
                                                Trace          =       5
Rotation: (unrotated = principal)          Rho             = 1.0000

--------------------------------------------------------------------------
   Component | Eigenvalue Difference       Proportion Cumulative
-------------+------------------------------------------------------------
         Comp1 |    2.86975    1.90278          0.5740    0.5740
         Comp2 |    .966968    .047566          0.1934    0.7673
         Comp3 |    .919402    .684109          0.1839    0.9512
         Comp4 |    .235294    .226709          0.0471    0.9983
         Comp5 | .00858493          .          0.0017    1.0000
--------------------------------------------------------------------------
然后我想要这样的结果

Eigenvalue

Proportion

Cumulative

2.539

50.782

1.199

23.975

74.758

1.018

20.365

95.122

请问可以直接用rotate吗？我试了fac ,然后再rotate还是没有弄出来求大神指教！谢谢！

13楼

饭团儿

发表于 2014-9-21 21:22:58

gg19861026 发表于 2014-9-21 17:32
你不需要做PCA，直接做factor就行，如果不会，请help factor

您好！我跟楼主的问是一样的用的命令也是一样就是先是pca后得到下面：

. pca liang e cang zhenfu jijinyijialv

Principal components/correlation                Number of obs =    13060
                                                Number of comp.  =       5
                                                Trace          =       5
Rotation: (unrotated = principal)          Rho             = 1.0000

--------------------------------------------------------------------------
   Component | Eigenvalue Difference       Proportion Cumulative
-------------+------------------------------------------------------------
         Comp1 |    2.86975    1.90278          0.5740    0.5740
         Comp2 |    .966968    .047566          0.1934    0.7673
         Comp3 |    .919402    .684109          0.1839    0.9512
         Comp4 |    .235294    .226709          0.0471    0.9983
         Comp5 | .00858493          .          0.0017    1.0000
--------------------------------------------------------------------------

Principal components (eigenvectors)

------------------------------------------------------------------------------
      Variable | Comp1    Comp2    Comp3    Comp4    Comp5 | Unexplained
-------------+--------------------------------------------------+-------------
         liang | 0.5761 0.0586 0.0830 -0.3772 -0.7180 |          0
            e | 0.5739 0.0401 0.0898 -0.4212 0.6954 |          0
         cang | 0.4673 0.5060 -0.0590 0.7220 0.0301 |          0
      zhenfu | 0.2854 -0.7849 0.3811 0.3965 0.0006 |          0
jijinyijialv |  -0.1974 0.3505 0.9145 -0.0431 -0.0014 |          0
------------------------------------------------------------------------------
有三个可作为主成分，
然后按您说的直接rotate之后得到：
. rotate

Principal components/correlation                Number of obs =    13060
                                                Number of comp.  =       5
                                                Trace          =       5
Rotation: orthogonal varimax (Kaiser off)    Rho             = 1.0000

--------------------------------------------------------------------------
   Component |    Variance Difference       Proportion Cumulative
-------------+------------------------------------------------------------
         Comp1 |          1  3.43547e-06          0.2000    0.2000
         Comp2 |          1  1.09934e-07          0.2000    0.4000
         Comp3 |          1 -7.89291e-08          0.2000    0.6000
         Comp4 |          1  4.29922e-06          0.2000    0.8000
         Comp5 |    .999996          .          0.2000    1.0000
--------------------------------------------------------------------------

Rotated components

------------------------------------------------------------------------------
      Variable | Comp1    Comp2    Comp3    Comp4    Comp5 | Unexplained
-------------+--------------------------------------------------+-------------
            e | 1.0000 0.0000 -0.0000 0.0000 -0.0000 |          0
         cang |  -0.0000 -0.0000 0.0000 1.0000 -0.0000 |          0
      zhenfu |  -0.0000 1.0000 -0.0000 0.0000 -0.0000 |          0
jijinyijialv | 0.0000 0.0000 1.0000 -0.0000 0.0000 |          0
         liang | 0.0000 0.0000 -0.0000 0.0000 1.0000 |          0
------------------------------------------------------------------------------

Component rotation matrix

----------------------------------------------------------------
               | Comp1    Comp2    Comp3    Comp4    Comp5
-------------+--------------------------------------------------
         Comp1 | 0.5739 0.2854 -0.1974 0.4673 0.5761
         Comp2 | 0.0401 -0.7849 0.3505 0.5060 0.0586
         Comp3 | 0.0898 0.3811 0.9145 -0.0590 0.0830
         Comp4 |  -0.4212 0.3965 -0.0431 0.7220 -0.3772
         Comp5 | 0.6954 0.0006 -0.0014 0.0301 -0.7180
-------------------------------------------------------------
然后我想问下通过上面的命令如何得到这个表格里面的结果，即降维后的系数呢？

旋转后方差特征值
总值       方差占比累计方差占比
2.539       50.782 50.782
1.199 23.975 74.758
1.018       20.365          95.122

stata新手，还请多多指教！非常感谢！