I want to find the depreciation using colSums - r

Question

I have a table (let's name it development.costs), as below:
1 2 3 4
4 5 6 7
8 9 1 2
Each column represents a year.
I want to create another table (let's name it depreciation.costs) with the same dimensions such that:
in each row, each element of each column is equal to:
0.4*[(Sum of all elements of that row of the development.costs table up until the year of the element) - (Sum of all elements of that row of the depreciation.costs table up until one year before)]
so I want to create a table
a b c d
e f g h
i j k l
such that e.g. c =0.4*[(1+2+3) - (a+b)]
the code I managed to write is
for (y in Years)
{depreciation.costs[y, ] <- 0.4*(colSums(development.costs[1:y], )-colSums(depreciation.costs[1:(y-1), ]))}
where Years <- 1:4
but this is wrong since the system gives me the error
Error in colSums(depreciation.rate[, 1:(y - 1)]) :
'x' must be an array of at least two dimensions
many thanks for any feedback

Answer 1

This seems to match the algorithm you describe, though it's difficult to tell from your description what the values in the first column are supposed to be.
Here's your data in matrix form:
dev_costs <- t(matrix(c(1:4, 4:7, 8:9, 1:2), nrow = 4))
dev_costs
#> [,1] [,2] [,3] [,4]
#> [1,] 1 2 3 4
#> [2,] 4 5 6 7
#> [3,] 8 9 1 2
We can easily make a cumulative sum of rows like this:
cum_dev <- t(apply(dev_costs, 1, cumsum))
Then an iterative loop to complete the algorithm:
answer <- cum_dev
for(i in seq(ncol(cum_dev))[-1])
{
answer[,i] <- 0.4 * (cum_dev[,i] - rowSums(answer[,1:(i-1), drop = FALSE]))
}
Giving us
answer
#> [,1] [,2] [,3] [,4]
#> [1,] 1 0.8 1.68 2.608
#> [2,] 4 2.0 3.60 4.960
#> [3,] 8 3.6 2.56 2.336
Created on 2020-03-06 by the reprex package (v0.3.0)