bootERGM: bootERGM: Exponential Random Graph Models by Parametric...

In schmid86/bootERGM: Exponential Random Graph Models with Parametric Bootstrapped Pseudolikelihood

Usage

bootERGM(formula, R=500, control.mple = NULL, control.sim = NULL, control.boot=NULL,
nowonit=100, cpu.cores=1, constraints =~.)

Arguments

formula	Formula for the TERGM. Model construction works like in the ergm package with the same model terms etc. (for a list of terms, see help("ergm-terms")).
R	The number of bootstrap replications. More replications lead to more accurate results, but slow down the estimation.
control.mple	ergm controls for ergm function in the ergm-package. Controls the ergm function that estimates the MPLE of the observed network. See control.ergm for details.
control.sim	simulation control for simulate.ergm function. See control.simulate.ergm for details.
control.boot	ergm controls for ergm function in the ergm-package. Controls the ergm-function that estimates the MPLE of the bootstrapped networks. See control.ergm for details.
nowonit	Returns: 'Now on Iteration nowonit' after every nowonit's iteration. Keeps track of the progress of the estimation. Default is nowonit=100.
cpu.cores	Number of CPU cores used. Default is cpu.cores=1. If the number of CPU cores used is greater than 1, the code will automatically parallelize the computation using the doParallel R-package.
constraints	Constraints of the ERGM. See ergm for details.

Author(s)

Christian Schmid ([e-mail])
Bruce Desmarais (http://brucedesmarais.com/)

References

C. S. Schmid and B. A. Desmarais. Exponential random graph models with big networks: Maximum pseudolikelihood estimation and the parametric bootstrap. 2017 IEEE International Conference on Big Data (Big Data), pages 116–121, Dec 2017

Examples

# load Faux Mesa High friendship data
data(faux.mesa.high)

# estimate model coefficients using parametric bootstrapped pseudolikelihood
m1 <- bootERGM(faux.mesa.high ~ edges + nodematch("Sex")+ gwesp(0.25, fixed=TRUE), R=500)
summary(m1)

schmid86/bootERGM documentation built on Nov. 5, 2019, 8:44 a.m.