VBLPCM (version 2.4.9)

vblpcmbic: calculate the BIC for the fitted VBLPCM object

Description

calculate the BIC for the fitted VBLPCM object

Usage

vblpcmbic(v.params)

Value

The scalar value of the BIC

Arguments

v.params

The fitted values; output from vblpcmfit()

Author

Michael Salter-Townshend

Details

BIC = BIC(edges | positions) + BIC(positions | clusters) w/ BIC(edges | positions) = -2 loglikelihood + (P+1)log(number of edges) and BIC(positions | clusters) as per mclust

References

Mark S. Handcock, Adrian E. Raftery and Jeremy Tantrum (2007). "Model-Based Clustering for Social Networks." Journal of the Royal Statistical Society: Series A (Statistics in Society), 170(2), 301-354.

See Also

latentnet::summary.ergmm

Examples

data(sampson)
set.seed(1)
### plot the BIC for G=2,3,4 groups 
gbic<-list(groups=NULL,bic=NULL)
for (g in 2:4)
  {
  v.fit<-vblpcmfit(vblpcmstart(samplike,G=g,LSTEPS=1e3),STEPS=20)
  gbic$groups[g]=v.fit$G
  gbic$bic[g]=v.fit$BIC$overall
  }
plot(gbic$groups, gbic$bic, main="BIC results", xlab="# groups", ylab="BIC", t='b')