why the coefficient of case 2 is 2 columns, not 5 columns? (about coefficient value of PCA)

Question

HYUN SEOK 2022-12-5

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1871357-why-the-coefficient-of-case-2-is-2-columns-not-5-columns-about-coefficient-value-of-pca

评论： HYUN SEOK 2022-12-7

(1) Case 1

[coeff] = pca(rand(5,3))
coeff = 3×3
    0.7120   -0.5694   -0.4109
    0.6460    0.3020    0.7010
    0.2750    0.7646   -0.5829

(2) Case 2

[coeff] = pca(rand(5,3)')
coeff = 5×2
   -0.0054    0.1255
   -0.4611    0.2098
   -0.1677    0.8905
   -0.5702   -0.3811
    0.6588    0.0448

why the coefficient of case 2 is 2 columns, not 5 columns?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Jon 2022-12-5

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1871357-why-the-coefficient-of-case-2-is-2-columns-not-5-columns-about-coefficient-value-of-pca#answer_1120697

编辑：Jon 2022-12-5

在 MATLAB Online 中打开

By default pca gives the 'economy' option which only includes significant components. for pca(X) with X n by p using 'economy' false will give a p by p coefficient matrix as you expected

 [coeff] = pca(rand(5,3)','economy',false)
coeff =
   -0.4049   -0.1882    0.5939   -0.5462    0.3867
   -0.5332    0.5469   -0.2904    0.2565    0.5162
    0.3748    0.7458    0.5429   -0.0101   -0.0925
   -0.1896   -0.2715    0.5169    0.7893   -0.0096
    0.6126   -0.1887   -0.0324    0.1128    0.7585