EpiModel (version 0.95)

epiNet.simNet: Simulate Networks from STERGM Model Fit

Description

This function simulates a set of dynamic networks of class networkDynamic given a STERGM model fit with epiNet.est.

Usage

epiNet.simNet(x, nsteps, nsims = 1, save.stats = TRUE, stats.formula,
  verbose = TRUE, ...)

Arguments

x

an EpiModel object of class epiNet.est.

nsteps

number of time steps over which to simulate relational dynamics.

nsims

number of network simulations.

save.stats

if TRUE, run simulation-based model diagnostics for the formation and dissolution processes.

stats.formula

a right-hand sided ERGM formula that includes network statistics of interest, with the default to the formation formula terms.

verbose

if TRUE, print simulation progress to the console.

...

additional arguments to pass to simulate.stergm.

Details

This function is a wrapper function around the simulate functions in the tergm package. The purpose is to simulate a series of dynamic relational networks for a set number of time steps, and to provide access to access to diagnostics on those network simulations.

The intended use of these simulated networks is in independent network disease models simulated in epiNet.simTrans. With dependent network models, one bypasses this network simulation step as that processes is coincident with the disease simulation. See Section 4 of ../doc/Tutorial.pdf{EpiModel Tutorial} for further clarification.

See Also

plot.epiNet.simNet, epiNet.est, epiNet.simTrans

Examples

# Initialize bipartite network
num.m1 <- 100
num.m2 <- 100
nw <- network.initialize(num.m1 + num.m2,
                         bipartite=num.m1, directed=FALSE)

# Set mode-specific fractional degree distributions
deg.dist.m1 <- c(0.40, 0.55, 0.04, 0.01)
deg.dist.m2 <- c(0.48, 0.41, 0.08, 0.03)

# Set model parameters and control arguments
formation <- ~ edges + b1degree(0:1) + b2degree(0:1)
target.stats <- c(66, 40, 55, 48, 41)
dissolution <- ~ offset(edges)
duration <- 25
coef.diss <- dissolution.coefs(dissolution, duration)

# Estimate the model with all the parameters set above
est <- epiNet.est(
  nw,
  formation,
  dissolution,
  target.stats,
  coef.diss)
est

# Simulate 5 networks over 100 timesteps
nwsims <- epiNet.simNet(est, nsteps = 100, nsims = 5)

# Print the object, plot the formation summary, plot the duration summary
nwsims
plot(nwsims, sim=2)
plot(nwsims, type = "duration", sim = 2)