This question already has an answer here:
- PCA and proportion of variance explained 4 answers
I've read through this explanation here regarding calculating the variance explained from PCA output. I think I got it right but might be off in my interpretation of R output.
In the example below, I would like to calculate the percentage of variance explained by the first principal component of the USArrests dataset.
pca <- prcomp(USArrests, scale = TRUE) eigs <- pca$sdev^2 eigs / sum(eigs)  0.6200604
I assumed that R uses
sdev as the square root of the eigen values. So I square it and divide the first value by the total. Is this correct?