PCA analysis: how can you identify which variable the first component corresponds to?

Question

I'm running PCA with 31 variables, and I need to know which variable corresponds to the firs principal component and captures the most variance. Here is my code:

mat=cov(df_15[11:41])
pca=princomp(covmat=mat)
summary(pca)

But the output names the components without mentioning any variables. Is there a way to find what I'm looking for? Am I misunderstanding it?

`loadings(pca)` will show you the loadings of the input variables on the components. The higher the loading of variable Vx on component Cx, the higher their correspondence. — I_O, Apr 30 '23 at 18:32
This is basically a statistics question and more suited for CV. — Rui Barradas, Apr 30 '23 at 18:42

score 1 · Answer 1 · answered Apr 30 '23 at 19:12

loadings(pca) shows the loadings of the input variables on the components. The higher the loading of variable Vx on component Cx, the higher their correspondence.

(As Rui Barradas mentioned, the mechanics of PCA would be better suited for Q&A on Cross Validated, see e. g. here.)

PCA analysis: how can you identify which variable the first component corresponds to?

1 Answers1