Confidence ellipses

The function ENlipse() allows to draw confidence ellipses around the categories of a chosen variable (for example a cluster variable).

Thanks to these ellipses, we can observe the degree of uncertainty of the points' position.

Here is an example using a cluster variable:

data(tea)
res.enmca=ENMCA(tea[,1:18])
x=res.enmca$MCA$ind$coord[,1:2]
b=res.enmca$data[,ncol(res.enmca$data)]
ENellipse(matP=x,cluster=b,resampling="population",eig=res.enmca$MCA$eig) #tea: the data set used whith the columns (i.e. the categorical variables) on which to perform the MCA
#matp: factorial coordinates from MCA
#cluster: chosen categorical variable
#resampling: If "population", the resampling is done within the whole dataset. If "cluster", it is done within the clusters
#eig: eigenvalues of the MCA

Scatterplot of the individuals coloured by clusterClick to viewConfidence ellipses around the clusters