NSBM.estimate {randnet}

estimates nomination SBM parameters given community labels by the method of moments

Description

estimates NSBM parameters given community labels

Usage

NSBM.estimate(A,K,g,reg.bound=-Inf)

Arguments

A

adjacency matrix of a directed where Aij = 1 iff i -> j

K

number of communities

g

a vector of community labels

reg.bound

the regularity lower bound of lambda value. By default, -Inf. That means, no constraints. When the network is sparse, using certain constaints may improve stability.

Details

The method of moments is used for estimating the edge nomination SBM, so the strategy can be used for both unweighted and weighted networks. The details can be found in Li et. al. (2020).

Value

a list of

B

estimated block connection probability matrix

lambda

estimated lambda values for nomination intensity

theta

estimated theta values for nomination preference

P.tilde

estimated composiste probability matrix after nomination

g

community labels

Author(s)

Tianxi Li, Elizaveta Levina, Ji Zhu
Maintainer: Tianxi Li [e-mail]

References

T. Li, E. Levina, and J. Zhu. Community models for networks observed through edge nominations. arXiv preprint arXiv:2008.03652 (2020).

See Also

SBM.estimate

Examples

dt <- NSBM.Gen(n=200,K=2,beta=0.2,avg.d=10)


A <- dt$A


sc <- RightSC(A,K=3)
est <- NSBM.estimate(A,K=3,g=sc$cluster)