CODEWITHSUNDEEP
raverage
Head and toes
Head and toes

Reputation: 659

Mean is not calculated correctly in R?

I am struggling to understand why this happens:

a<-c(1.1,1.2,1.3)
b<-c(2.1,2.3,2.6)
c<-c(1.6,2.3,2.6)

mean<-rowMeans(matrix(c(a,b,c),ncol=3))  
mean  #### This are the mean values
[1] 1.600000 1.933333 2.166667

mean(a[1],b[1],c[1])  #### trying to calculate the mean of 1.1 ,2.1 and 1.6
[1] 1.1  #### Why is this not 1.6??

Upvotes: 1

Views: 1027

Answers (3)

Piotrek
Piotrek

Reputation: 3

Mabye your problem is that your matrix is transposed? Try this:

mean<-rowMeans(matrix(c(a,b,c),ncol=3, byrow=T))

Upvotes: 0

Spacedman
Spacedman

Reputation: 94317

R's argument matching has caught you out:

> mean(9999,2,3,4,5)
[1] 9999

help(mean) says:

Usage:

     mean(x, ...)

and then says it computes the mean of x and passes ... to other methods. All the numbers after 9999 in my example get captured in the dot-dot-dot.

R then calls mean.default because 9999 is just a number, and mean.default doesn't do anything with the dot-dot-dot arguments, including erroring if there's anything in them.

You can use this to add arbitrary useless things to your function calls:

> mean(c(1,2,3,4), monkeys=TRUE)
[1] 2.5

Upvotes: 14

nrussell
nrussell

Reputation: 18612

Your call to mean in mean(a[1],b[1],c[1]) is only operating on the first element, i.e. mean(a[1]). You need to concatenate a[1],b[1],c[1] like this: 

> mean(c(a[1],b[1],c[1]))
[1] 1.6

Upvotes: 5

Related Questions