Principal Component Analysis
Daniel J. Denis · 2020
Principal component analysis (PCA) is a data reduction technique used to account for variance in a set of p variables into fewer than p dimensions. PCA is usually applied to data sets that contain many variables, sometimes hundreds, since it is for these data sets especially that we wish to reduce dimensionality. One example of a PCA is actually one featured by a pioneer of component analysis, Karl Pearson, infamous statistician who first introduced the technique in a paper written in 1901. This chapter explains how to run and interpret a component analysis in R using prcomp() and princomp() functions. PCA can be performed on either the original unstandardized data, or on data standardized to have mean equal to zero and standard deviation equal to 1. This transformation is equivalent to analyzing the correlation matrix (standardized data) rather than the covariance matrix.
Cited by 6 works. doi:10.1002/9781119549963.ch10