Skip to content

Commonly used ERGM terms

Martina Morris edited this page Jul 29, 2021 · 1 revision

For full list of the terms available in the ergm package, in R type: help("ergm-terms")

Name Unit of counting Description Num. of statistics Dep/ind Dir/undir Unip/bip
edges edges number of edges 1 ind both both
nodefactor units of the attribute number of times that nodes with a given level of a categorical nodal attribute appear within the edgeset 1 per level of attribute included ind both both
nodematch edges number of edges whose incident nodes match on value of nodal attribute if diff=F, 1; if diff=T, 1 per level of attribute included ind both both
nodemix edges number of edges whose incident nodes fall into each combination of values for a nodal attribute 1 per combo of attribute included ind both both
nodecov units of the attribute the sum of a nodal attribute value for the incident nodes of an edge, summed across all edges 1 ind both both
absdiff units of the attribute the absolute difference in a nodal attribute between incident nodes for an edge, summed across all edges 1 ind both both
mutual dyad number of mutual edges (a->b and b->a) 1 dep dir both
degree node number of nodes of degree x 1 per each value of x included dep undir unip
idegree node number of nodes of in-degree x 1 per each value of x included dep dir unip
odegree node number of nodes of out-degree x 1 per each value of x included dep dir unip
b1degree node number of mode 1 nodes of degree x 1 per each value of x included dep undir bip
b2degree node number of mode 2 nodes of degree x 1 per each value of x included dep undir bip
mindegree node number of nodes of at least degree x 1 per each value of x included dep undir unip
triangles triangles number of triangles 1 (more if attributes used) dep undir unip
gwesp complex geometrically weighted edgewise shared partners 1 dep both unip

Notes:

  • Dep/indep = A term involving statistical dependence or independence among relations

  • Dir/undir = works for directed and/or indirected networks

  • Unip/bip = works for unipartite and.or bipartite networks