Factoextra: visual PCA of multivariate statistical methods

FactoMineR and factoextra The dimension reduction methods included can be based on the characteristics of data variables （ ration / The number 、 qualitative / classification 、 blend ） Distinguish between these methods .

Principal component analysis （PCA）

example ： Los Angeles block data

From the U.S. Census Bureau 2000 Years of data , Including the Los Angeles area 110 A block ,15 A variable . Use rvest package , Get this data set from the network . its 15 The variables are shown in the following table ：

surface ： Los Angeles block dataset

We only use six of these variables （Income,Schools,Diversity,Age,Homes,Vets） Principal component analysis , Among the remaining variables, take out four variables related to race （Asian,Black,Latino,White） As a supplementary variable for exploratory data analysis . Use FactoMineR Function of the package PCA() Realize principal component analysis ：

library(FactoMineR)
library(factoextra)
library(rvest)
<