# [R] LDA and confidence ellipse

arun smartpink111 at yahoo.com
Thu Jul 11 20:25:19 CEST 2013

```Hi,
May be this helps:
require(MASS)
require(ggplot2)
iris.lda<-lda(Species ~ Sepal.Length + Sepal.Width + Petal.Length + Petal.Width,  data = iris)
datPred<-data.frame(Species=predict(iris.lda)\$class,predict(iris.lda)\$x)

library(ellipse)
dat_ell <- data.frame()
for(g in levels(datPred\$Species)){
dat_ell <- rbind(dat_ell, cbind(as.data.frame(with(datPred[datPred\$Species==g,], ellipse(cor(LD1, LD2),
scale=c(sd(LD1),sd(LD2)),
centre=c(mean(LD1),mean(LD2))))),Species=g))
}

ggplot(datPred, aes(x=LD1, y=LD2, col=Species) ) + geom_point( size = 4, aes(color = Species))+theme_bw()+geom_path(data=dat_ell,aes(x=x,y=y,color=Species),size=1,linetype=2)

A.K.

Hi,

I wish to add confidence ellipse on my LDA result of the iris data set.
Therefore:
Is there statistical logic to do that as I only wish it to make the species separation more visable?
How can I add it to the script below  (ggplot):
require(MASS)
require(ggplot2)
iris.lda<-lda(Species ~ Sepal.Length + Sepal.Width + Petal.Length + Petal.Width,  data = iris)
LD1<-predict(iris.lda)\$x[,1]
LD2<-predict(iris.lda)\$x[,2]
ggplot(iris, aes(x=LD1, y=LD2, col=iris\$Species) ) + geom_point( size = 4, aes(color = iris\$Species))+theme_bw()

Could someone please help me. Thank you very much.

```

More information about the R-help mailing list