You need first to find the indices of all those elements within 2 standard deviations of the vector. Then you can use those indices to select such elements.
Here is an example:
> aa <- c(10,12,324,13,345,32,56,73,87,19,23,45,67, 20,42,120, 112)> pos = abs(aa - mean(aa)) <= 2*sd(aa)> pos [1] TRUE TRUE FALSE TRUE FALSE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE> bb <- aa[pos]> bb [1] 10 12 13 32 56 73 87 19 23 45 67 20 42 120 112>
If you want to use one line code, you can combine two statements like this:
> bb <- aa[abs(aa - mean(aa)) <= 2*sd(aa)]> bb [1] 10 12 13 32 56 73 87 19 23 45 67 20 42 120 112