The package “corrplot” is used for graphical display of a correlation matrix. The features in this package can be used to customise the correlation matrix plot. We can choose text labels, colour labels, layout, etc. Also there are options available for matrix reordering.
Install and load package corrplot
install.packages("corrplot")
library(corrplot)
MICA<-read.csv("Motor Insurance claim amount.csv",header=TRUE)
head(MICA)
vehage CC Length Weight claimamt
1 4 1495 4250 1023 72000.0
2 2 1061 3495 875 72000.0
3 2 1405 3675 980 50400.0
4 7 1298 4090 930 39960.0
5 2 1495 4250 1023 106800.0
6 1 1086 3565 854 69592.8
c<-cor(MICA)
c
vehage CC Length Weight claimamt
vehage 1.0000000 0.1282000 0.1882112 0.1402555 -0.6009692
CC 0.1282000 1.0000000 0.8041008 0.9055685 0.4580953
Length 0.1882112 0.8041008 1.0000000 0.8264651 0.4780824
Weight 0.1402555 0.9055685 0.8264651 1.0000000 0.4169866
claimamt -0.6009692 0.4580953 0.4780824 0.4169866 1.0000000
corrplot(c,method="circle")
Positive correlations are displayed in blue and negative correlations in red color. Color intensity and the size of the circle are proportional to the correlation coefficients.
There are seven visualization methods in corrplot package, namely circle, square, ellipse, number, shade, color and pie
corrplot(c,method="number")
There are three types of layout
full (default) : display full correlation matrix
upper : display upper triangular of the correlation matrix
lower : display lower triangular of the correlation matrix
corrplot(c,method = "square", type = "upper")
corrplot.mixed()
is a function used to display mixed visualization style.
lower=
specifies the visualization method for the lower triangular correlation matrix.
upper=
specifies the visualization method for the upper triangular correlation matrix.
corrplot.mixed(c,lower = "number", upper = "ellipse" )
We can also specify the colour and font size of the number using lower.col=
and number.cex=
corrplot.mixed(c, lower.col = "black", number.cex = .7)
The correlation matrix can be reordered according to the correlation coefficient. This is important to identify the hidden structure and pattern in the matrix.
There are four methods for reordering in corrplot
AOE - based on angular order of the eigenvectors
FPC - based on first principal component order
hclust - based on hierarchical clustering of the correlation matrix
alphabet - based on alphabetical order
corrplot(c, order = "hclust")