IQR outlier in R

Question

I am supposed to use the 1.5*IQR rule to determine outliers on the left and right tail by using these two equations in a function:

• Q1-(1.5*IQR)
• Q3+(1.5*IQR)

This is what I have tried so far:

 IQR.outliers <- function(x) { if(any(is.na(x))) stop("x is missing values") if(!is.numeric(x)) stop("x is not numeric") Q3<-quantile(x,0.75) Q1<-quantile(x,0.25) IQR<-(Q3-Q1) left<-x<(Q1-(1.5*IQR)) right<-x>(Q3+(1.5*IQR)) c(left,right) } 

Answer 1

At the moment you are getting a vector of logicals. Instead use logical indexing. Assuming you just want the values themselves and not their positions, then:

IQR.outliers <- function(x) {
  if(any(is.na(x)))
    stop("x is missing values")
  if(!is.numeric(x))
    stop("x is not numeric")
  Q3<-quantile(x,0.75)
  Q1<-quantile(x,0.25)
  IQR<-(Q3-Q1)
  left<- (Q1-(1.5*IQR))
  right<- (Q3+(1.5*IQR))
  c(x[x <left],x[x>right])
}

Check to see if it gives sensible results:

> IQR.outliers (1:100)
integer(0)
> IQR.outliers (c(1:100,1000))
[1] 1000
> IQR.outliers (rnorm(1000) )
[1] -3.522064 -3.296682  2.910200  2.671333