ergm package
that links the graphlet orbits of a given type with categorical node attributes.
grorbitFactor(attrname, grorbit, base)grorbitFactor term adds a relationship between graphlet degrees and a
categorical node attribute into an ERGM. The question that the change score function
of this term answers is: what is the change in the total graphlet degree (for a given
orbit) for those nodes with a given attribute value, for a particular edge change?
This term has three arguments: attrname; grorbit; and base.
attrname is a character vector giving the name of a categorical attribute
in the network's vertex attribute list. The optional grorbit argument is a
vector of distinct integers representing the list of graphlet orbits to include
into the model. When grorbit is not provided, all graphlet orbits are evaluated
by default. The optional base argument is a vector of distinct integers representing
the list of categories in attrname that are going to be omitted. When this
argument is set to 0, all categories are evaluated. When this argument is set to 1,
the category having the lowest value (or lexicographically first name) is eliminated.
The term sorts all values of the categorical attribute lexicographically and
base term defines the indexes of the categories to be omitted in this sorted
list. For example, if the "fruit" attribute has values "orange", "apple", "banana"
and "pear", grorbitFactor("fruit" , 0 , 2:3) will ignore the "banana" and
"orange" factors and evaluate the "apple" and "pear" factors. When the base
argument is not provided, the argument is set to 1 by default. The
grorbitFactor term adds a x |grorbit| terms into the model where
a represents the number of attribute values that are evaluated in the model
and |grorbit| is the number of graphlet orbits to be evaluated in the model.
This term is defined for the 73 graphlet orbits corresponding to graphlets with
up to 5 nodes. Therefore, grorbit accepts values between 0 and 72. Values
outside this range are ignored. This term is defined for undirected networks.
ergm is used to fit linear
exponential random graph models, in which the
probability of a given network, $y$, on a set of nodes
is $\exp{\theta *
g(y)}/c(\theta)$, where $g(y)$ is a vector of network
statistics for $y$, $\theta$ is a parameter vector of the same
length and $c(\theta)$ is the normalizing constant for the distribution.The network statistics $g(y)$ are entered as terms in the
function call to ergm. The ergm package contains
a wide range of terms. For the details on the possible
ergm-terms.
The ergm.userterms package provides a template for
adding new terms. The terms can be used throughout the ergm package
and behave identically to the supplied terms.