For each of my N variables, I have a (T * M) feature matrix, i.e., M observations per t \in T. The problem is how to convert this into a (T * N * M) array. For example, in the following example N=2, T=3, M=4 :
x1 <- matrix(1:24, 3,4)
> x1
[,1] [,2] [,3] [,4]
[1,] 1 4 7 10
[2,] 2 5 8 11
[3,] 3 6 9 12
x2 <- matrix(25:48, 3,4)
x2
[,1] [,2] [,3] [,4]
[1,] 25 28 31 34
[2,] 26 29 32 35
[3,] 27 30 33 36
And I need to make a 3 dimensional (number of rows) array, such that the first element is
[,1] [,2] [,3] [,4]
[1,] 1 4 7 10
[2,] 25 28 31 34
and the second is:
[,1] [,2] [,3] [,4]
[1,] 2 5 8 11
[2,] 26 29 32 35
and third:
[,1] [,2] [,3] [,4]
[1,] 3 6 9 12
[2,] 27 30 33 36
and so on and so forth. For the following example, the output's dimensions should be (3,2,4).
I need to do this for relatively large N and T, so appreciate extendable implementations!
Here is a base R option.
out <- `dim<-`(rbind(c(t(x1)), c(t(x2))), c(2, 4, 3))
out
#, , 1
#
# [,1] [,2] [,3] [,4]
#[1,] 1 4 7 10
#[2,] 25 28 31 34
#
#, , 2
#
# [,1] [,2] [,3] [,4]
#[1,] 2 5 8 11
#[2,] 26 29 32 35
#
#, , 3
#
# [,1] [,2] [,3] [,4]
#[1,] 3 6 9 12
#[2,] 27 30 33 36
When we call x <- rbind(c(t(x1)), c(t(x2))) we get the following matrix as a result
x
# [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
#[1,] 1 4 7 10 2 5 8 11 3 6 9 12
#[2,] 25 28 31 34 26 29 32 35 27 30 33 36
We need to change the dimensions of this object for which we can do
dim(x) <- c(2, 4, 3)
Another way to get the same result is to call the replacement method of dim in its functional form, i.e.
`dim<-`(...)
Which allows us to do all in one line.