[Stata初级班] 问一个很初级的关于factor analysis的问题。。。 [推广有奖]

0关注
1粉丝

小学生

92%

还不是VIP/贵宾

-

0%

威望: 0 级
论坛币: 10 个
通用积分: 0
学术水平: 0 点
热心指数: 0 点
信用等级: 0 点
经验: 66 点
帖子: 6
精华: 0
在线时间: 5 小时
注册时间: 2012-3-21
最后登录: 2012-5-21

楼主

rena8822 发表于 2012-4-12 17:20:21 |AI写论文

是否 +2 论坛币

k人参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

我想知道几个问题能不能用来measure一个variable。
factor analysis的结果是

--------------------------------------------------------------------------
Factor    Eigenvalue Difference       Proportion Cumulative
-------------+------------------------------------------------------------
Factor1       4.17977    3.34728          0.9516    0.9516
Factor2       0.83249    0.80832          0.1895    1.1411
Factor3       0.02416    0.03348          0.0055    1.1466
Factor4    -0.00932    0.04828          -0.0021    1.1445
Factor5    -0.05760    0.00238          -0.0131    1.1314
Factor6    -0.05998    0.05057          -0.0137    1.1177
Factor7    -0.11055    0.01519          -0.0252    1.0925
Factor8    -0.12574    0.00724          -0.0286    1.0639
Factor9    -0.13297    0.01477          -0.0303    1.0336
Factor10    -0.14775          .          -0.0336    1.0000
--------------------------------------------------------------------------
LR test: independent vs. saturated:  chi2(45) = 9.2e+04 Prob>chi2 = 0.0000
Factor loadings (pattern matrix) and unique variances
-----------------------------------------------------------
Variable Factor1 Factor2 Factor3 Uniqueness
-------------+------------------------------+--------------
a8d 0.5476 0.3270 0.0322    0.5921
a8e 0.5838 0.3287 0.0358    0.5499
a8f 0.4762 0.2980 0.0260    0.6838
a8g 0.6444 0.3554 -0.0331    0.4574
a7a 0.6822 -0.1993 -0.0726    0.4897
a7b 0.7161 -0.2563 -0.0392    0.4200
a7c 0.7326 -0.3071 -0.0175    0.3687
a7d 0.7363 -0.2978 0.0532    0.3663
a7e 0.6294 -0.2284 0.0873    0.5440
a8b 0.6664 0.2486 -0.0500    0.4917
-----------------------------------------------------------
rotate以后是
Factor analysis/correlation                      Number of obs = 22451
Method: principal factors                   Retained factors =       3
Rotation: orthogonal varimax (Horst off)    Number of params =    27
--------------------------------------------------------------------------
      Factor  |    Variance Difference       Proportion Cumulative
-------------+------------------------------------------------------------
      Factor1  |    2.90073    0.78948          0.6604    0.6604
      Factor2  |    2.11125    2.08682          0.4806    1.1410
      Factor3  |    0.02443          .          0.0056    1.1466
--------------------------------------------------------------------------
LR test: independent vs. saturated:  chi2(45) = 9.2e+04 Prob>chi2 = 0.0000
Rotated factor loadings (pattern matrix) and unique variances
-----------------------------------------------------------
      Variable |  Factor1 Factor2 Factor3 | Uniqueness
-------------+------------------------------+--------------
         a8d | 0.2283 0.5952 0.0391 |    0.5921
         a8e | 0.2557 0.6188 0.0430 |    0.5499
         a8f | 0.1901 0.5283 0.0321 |    0.6838
         a8g | 0.2868 0.6780 -0.0253 |    0.4574
         a7a | 0.6594 0.2659 -0.0699 |    0.4897
         a7b | 0.7213 0.2416 -0.0369 |    0.4200
         a7c | 0.7657 0.2117 -0.0156 |    0.3687
         a7d | 0.7629 0.2205 0.0553 |    0.3663
         a7e | 0.6360 0.2084 0.0894 |    0.5440
         a8b | 0.3701 0.6079 -0.0430 |    0.4917
-----------------------------------------------------------
Factor rotation matrix
-----------------------------------------
               | Factor1  Factor2  Factor3
-------------+---------------------------
      Factor1 |  0.7861 0.6181 0.0068
      Factor2 | -0.6182 0.7860 0.0098
      Factor3 |  0.0007  -0.0119 0.9999
-----------------------------------------
所以，我到底是要用rotate之前的结果还是rotate之后的啊？那这几个能不能放在一块儿用来measure一个variable 啊？
可以的话，接下来是不是直接用predict的命令就可以了？