steps_ERE: Perform Steps of the Gibbs Sampler of the ERE model

In systemicrisk: Systemic Risk and Network Reconstruction

Perform Steps of the Gibbs Sampler of the ERE model

Description

Runs a Gibbs sampler in the Erdos Reny model with Exponential weights (ERE model) and fixed marginals. The algorithm starts from a given matrix.

Usage

steps_ERE(L, p, lambda, nsamples = 10000, thin = 1000, burnin = 10000)

Arguments

L	Starting matrix for the Gibbs sampler. Implicitly defines the fixed marginals.
p	A matrix with entries in [0,1]
lambda	A matrix with nonnegative entries
nsamples	Number of samples to return.
thin	Frequency at which samples should be generated (default=1, every step)
burnin	Number of initial steps to discard.

Value

List of simulation results

See Also

sample_ERE

Examples

L <- matrix(rexp(4*4),nrow=4,ncol=4); diag(L)=0;
p <- matrix(0.5,nrow=4,ncol=4); diag(p) <-0;
lambda <- matrix(1,nrow=4,ncol=4); diag(lambda)<-0;

L <- steps_ERE(L=L,p=p,lambda=lambda,nsamples=5,thin=50,burnin=20)
L

systemicrisk documentation built on May 29, 2024, 9:20 a.m.