This term adds one statistic to the model, equal to the number of cross-layer two-stars or two-paths in the network.
# binary: twostarL(Ls, type, distinct=TRUE)a list (constructed by list() or c() of two Layer Logic specifications (c.f. Layer Logic section in the Layer() documentation) specifying the layers of interest.
if the network is directed, an argument to determine which configurations are counted:
"out"Number of configurations
\((i{\rightarrow}j), (i{\rightarrow}k)\), where
\((i{\rightarrow}j)\) is in logical layer Ls[[1]]
and \((i{\rightarrow}k)\) is in logical layer Ls[[2]].
"in"Number of configurations
\((j{\rightarrow}i), (k{\rightarrow}i)\), where
\((j{\rightarrow}i)\) is in logical layer Ls[[1]]
and \((k{\rightarrow}i)\) is in logical layer Ls[[2]].
"path"Number of configurations
\((j{\rightarrow}i), (i{\rightarrow}k)\), where
\((j{\rightarrow}i)\) is in logical layer Ls[[1]]
and \((i{\rightarrow}k)\) is in logical layer Ls[[2]].
This argument is ignored for undirected networks.
if TRUE, \(j\) and \(k\) above are required to
be distinct. That is, the constituent edges may not be coincident or
reciprocal.
ergmTerm for index of model terms currently visible to the package.
ergm:::.formatTermKeywords("ergmTerm", "twostarL", "subsection")