ergm.bridge.llr uses bridge sampling with geometric spacing to
estimate the difference between the log-likelihoods of
two parameter vectors for an ERGM via repeated calls to
simulate.formula.ergm.
ergm.bridge.0.llk is a convenience wrapper around ergm.bridge.llr: returns the
log-likelihood of configuration `theta' relative to the reference
measure. That is, the configuration with theta=0 is defined as
having log-likelihood of 0
See also ergm.bridge.dindstart.llk to use dyad-independent ERGM as a staring point.
ergm.bridge.llr(object,
response=NULL,
constraints=~.,
from,
to,
basis=NULL,
verbose=FALSE,
…,
llronly=FALSE,
control=control.ergm.bridge())
ergm.bridge.0.llk(object,
response=response,
coef,
...,
llkonly=TRUE,
control=control.ergm.bridge())A model formula. See ergm for details.
Not for release.
A one-sided formula specifying one or more constraints
on the support of the distribution of the networks being
simulated. See the documentation for a similar argument for
ergm for more information. For
simulate.formula, defaults to no constraints. For
simulate.ergm, defaults to using the same constraints as
those with which object was fitted.
The initial and final parameter vectors.
An optional network object to start
the Markov chain. If omitted, the default is the left-hand-side of
the object.
Logical: If TRUE, print detailed information.
Further arguments to ergm.bridge.llr and simulate.formula.ergm.
Logical: If TRUE, only the estiamted log-ratio will be returned.
Control arguments. See
control.ergm.bridge for details.
A vector of coefficients for the configuration of interest.
Whether only the estiamted log-likelihood should be returned. (Defaults to TRUE.)
If llronly=TRUE, returns the scalar
log-likelihood-ratio. Otherwise, returns a list with the following components:
The estimated log-ratio.
The estimated log-ratios for each of the nsteps
bridges.
A numeric matrix with nsteps rows, with each row being the respective bridge's parameter configuration.
A numeric matrix with nsteps rows, with each row being the respective bridge's vector of simulated statistics.
The gradient vector of the parameter values with respect to position of the bridge.
Hunter, D. R. and Handcock, M. S. (2006) Inference in curved exponential family models for networks, Journal of Computational and Graphical Statistics.