R: calculating mean for every n different values from a vector

Question

I want to calculate the mean of the 1st value to the 10th, then calculate the mean for the 11th value to the 15th and then calculate the mean for the 16th value to the 18th and so. So basically I want to calculate the mean for the Xth values where X are different values given in a vector. For example:

y=c(1,1,2,3,2,1,4,5,3,6,7,5,6,7,8,4,2,4)
x=c(10,5,3)

Then, the first, second and third means would be:

sum(1+1+2+3+2+1+4+5+3+6)/10
sum(7+5+6+7+8)/5
sum(4+2+4)/3

Can you help me please? Thanks!

`start = 1; for (v in cumsum(x)) { print(mean(y[start:v])); start = v + 1 }` — Abdou, Nov 22 '17 at 05:18

score 3 · Answer 1 · answered Nov 22 '17 at 05:19

3

You can borrow the splitAt function from this answer:

splitAt <- function(x, pos) {
  unname(split(x, cumsum(seq_along(x) %in% pos)))
}

To split y you need to start at 1 and add 1 to the cumulative sum of the values in x:

splitAt(y, c(1, cumsum(x)+1))

[[1]]
 [1] 1 1 2 3 2 1 4 5 3 6

[[2]]
[1] 7 5 6 7 8

[[3]]
[1] 4 2 4

Now you have a list, so you can lapply or sapply:

sapply(splitAt(y, c(1, cumsum(x)+1)), mean)

[1] 2.800000 6.600000 3.333333

answered Nov 22 '17 at 05:19

neilfws

32,751
5
50
63

Yes, I feel like the `splitAt` function could be modified to make the `sapply` procedure more elegant. Anyway, seems to work OK as is. – neilfws Nov 22 '17 at 05:33
Thank you so much ! neilfws – David Vivas Dec 09 '17 at 05:31

score 1 · Accepted Answer · edited Nov 22 '17 at 05:34

1

x <- cumsum(x)
x <- c(1,x)
for(i in 1:(length(x)+1)){
    print(mean(y[x[i-]:x[i+1]]))

edited Nov 22 '17 at 05:34

neilfws

32,751
5
50
63

answered Nov 22 '17 at 05:09

Tomo

580
6
5

R: calculating mean for every n different values from a vector

2 Answers2