I want to create a design matrix for use in the glmnet function. My predictors are player names; when player i plays a match against player j I want the row for that match to have a +1 in column i and a -1 in column j.
For example, if there are three players who all play each other, the design matrix should be as follows:
1 -1 0
1 0 -1
0 1 -1
I've tried using model.matrix but I'm not sure how to create this specific design matrix - alternatively if there's a better format for this matrix that would also be helpful to know!
Try this:
num_players <- 4
matches <- t(combn(num_players, 2))
nr <- nrow(matches)
mtx <- matrix(0L, nrow = nr, ncol = num_players)
mtx[ cbind(seq_len(nr), matches[,1]) ] <- 1L
mtx[ cbind(seq_len(nr), matches[,2]) ] <- -1L
mtx
# [,1] [,2] [,3] [,4]
# [1,] 1 -1 0 0
# [2,] 1 0 -1 0
# [3,] 1 0 0 -1
# [4,] 0 1 -1 0
# [5,] 0 1 0 -1
# [6,] 0 0 1 -1