nbOrder: Determine Neighbourhood Order Matrix from Binary Adjacency...
In jimhester/surveillance: Temporal and Spatio-Temporal Modeling and Monitoring of Epidemic Phenomena

Description Usage Arguments Value Note Author(s) See Also Examples

Given a square binary adjacency matrix, the function nbOrder determines the integer matrix of neighbourhood orders (shortest-path distance) using the function nblag from the spdep package.

1	nbOrder(neighbourhood, maxlag = 1)

`neighbourhood`	a square, numeric or logical, and usually symmetric matrix with finite entries (and usually zeros on the diagonal) which indicates vertex adjacencies, i.e., first-order neighbourhood (interpreted as `neighbourhood == 1`, not `>0`).
`maxlag`	positive scalar integer specifying an upper bound for the neighbourhood order. The default (1) just returns the input neighbourhood matrix (converted to binary integer mode). `maxlag` is automatically trimmed to one less than the number of regions (there cannot be higher orders) and then converted to integer, thus, `maxlag = Inf` also works.

An integer matrix of neighbourhood orders, i.e., the shortest-path distance matrix of the vertices. The dimnames of the input neighbourhood matrix are preserved.

By the end, the function issues a message informing about the range of maximum neighbourhood order by region.

Sebastian Meyer

nblag from the spdep package, on which this wrapper depends.

## generate adjacency matrix
set.seed(1)
n <- 6
adjmat <- matrix(0, n, n)
adjmat[lower.tri(adjmat)] <- sample(0:1, n*(n-1)/2, replace=TRUE)
adjmat <- adjmat + t(adjmat)
adjmat

## determine neighbourhood order matrix
if (requireNamespace("spdep")) {
    nbmat <- nbOrder(adjmat, maxlag=Inf)
    nbmat
}