Transforms the adjacency matrix of a graph into
an ``edge matrix''.
Usage
edgematrix(E, inv=FALSE)
Arguments
E
a square matrix, representing the adjacency matrix of a
graph.
inv
a logical value.
Value
Athe edge matrix of the graph.
If TRUE the nodes are sorted in
inverted topological order and the edge matrix is upper triangular.
Details
In some matrix computations for graph objects the adjacency matrix
of the graph is transformed into an ``edge matrix''. Briefly,
if $E$ is the adjacency matrix of the
graph, the edge matrix is $A = sign(E+I)^T=[a_{ij}]$.
Thus, $A$ has ones along the diagonal
and if the graph has no edge between nodes $i$ and $j$ the entries
$a_{i,j}$ and $a_{j,i}$ are both zero.
If there is an arrow from $j$ to $i$
$a_{i,j}=1$ and $a_{j,i} = 0$. If there is an undirected edge, both
$a_{i,j}=a_{j,i}=1$.
References
Wermuth, N. (2003). Analysing social science data with
graphical Markov models. In: Highly Structured Stochastic
Systems. P. Green, N. Hjort & T. Richardson (eds.),
47--52. Oxford: Oxford University Press.