r - How to visualize PCA variable loadings as barplot? - Stack Overflow

admin管理员组

文章数量:1344303

I am currently using R Studio and tackling a task that'd require me to create a visual that allows me to examine the loadings from PC1 and PC2 from a PCA and identify which loadings (disregarding the sign) are greater than 0.3.

I am aware that this information is readily and simply available by looking at the console output for:

round(pca$rotation, 3)

Where I am rounding to 3 d.p. the Loadings/coefficients for the PCs found in a PCA called "pca"

However, for the sake of my task, I'd like to be able to reinforce these numbers with a simple visual like a barplot or something that's doable using tidyverse packages.

I am currently using R Studio and tackling a task that'd require me to create a visual that allows me to examine the loadings from PC1 and PC2 from a PCA and identify which loadings (disregarding the sign) are greater than 0.3.

I am aware that this information is readily and simply available by looking at the console output for:

round(pca$rotation, 3)

Where I am rounding to 3 d.p. the Loadings/coefficients for the PCs found in a PCA called "pca"

However, for the sake of my task, I'd like to be able to reinforce these numbers with a simple visual like a barplot or something that's doable using tidyverse packages.

• r
• pca

Share Improve this question Follow edited yesterday Xinovy asked yesterday XinovyXinovy 1133 bronze badges New contributor Xinovy is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct.

Add a comment  | 

1 Answer 1

Sorted by: Reset to default 1

Just use barplot.

barplot(pcar[, c('PC1', 'PC2')], beside=TRUE, col=seq_len(ncol(pcar)) + 1,
        legend=rownames(pcar), args.legend=list(x='top'))
abline(h=0.3, col='red', lty=2)

---

Data:

pcar <- prcomp(USArrests)$rotation

本文标签： rHow to visualize PCA variable loadings as barplotStack Overflow