Aug 20, 2023

PCA does not indicate the relative importance of each variable. It shows, when the data is recast onto a new domain to want extent the variance is described by these components.

The plot above shows that one new component accounts for 62% of the variance. The plot your commenting on doesn't state relative importance in the way you described it nor is one of the numbers 50%.

The plot your commenting on looks like the correlation coefficient between the principal component and the original features.