stata中主成分分析法中不理解的命令、符号

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

我是通过初始变量的协方差矩阵进行主成分分析，最终保留2个主成分，主成分分析的结构就如图中所示，请问图中所说的trace、difference、unexplained是什么意思？我知道英文翻译过来时踪迹、不同值、未经解释的，但是放到stata中到底是代表什么含义？求好心人帮帮忙，我是初学者，木有金币，真心感激好心人！

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏3 回帖

关键词：主成分分析法 主成分分析 Stata tata 主成分 分析法

有木有人能帮忙解决我的难题？求好心人！！！

trace is the trace of the covariance matrix or the total variance.  Difference is the difference between the eigenvalues (component1 eigenvalue - component2 eigenvalue, and so on). unexplained is the unexplained variances caused by omitting the other components.  You may want to read the principle component section to get better idea as what it does and what the results mean.

jjjj6666 发表于 2014-2-25 01:49
trace is the trace of the covariance matrix or the total variance.  Difference is the difference bet ...

谢谢您的帮忙，但是“trace is the trace of the covariance matrix or the total variance”这句话能麻烦您翻译一下吗？这里的trace是踪迹的意思，踪迹是协方差或者是方差的踪迹，我翻译成这样，应该不对吧？

suppose the covariance matrix is (3,2\2,4), the trace is 3+4=7.  This is also the total variance: if you sum the variances of the variables "trunk weight length headroom", it is the same as the number in trace.

jjjj6666 发表于 2014-2-25 11:06
suppose the covariance matrix is (3,2\2,4), the trace is 3+4=7.  This is also the total variance: if ...

原来是这样，谢谢您！