Split a matrix into block diagonal sub matrices according to clusters and
combine the lower triangular parts into a vector
lower.tri.vector(x, cluster = rep(1, nrow(x)), diag = FALSE)
Arguments
- x
a square matrix
- cluster
numeric or factor. Is used to identify the sub-matrices of
x from which the lower triangular parts are extracted. Defaults to
the full matrix.
- diag
logical. Should the diagonal be included?
Value
Returns a numeric vector containing the elements of the lower
triangular sub matrices.
Examples
m <- matrix(1:64, ncol=8)
cluster <- c(1, 1, 1, 1, 2, 2, 3, 3)
lower.tri.vector(m, cluster)
#> [1] 2 3 4 11 12 20 38 56