cv.SBA {graphon}R Documentation

Cross validation for selecting optimal precision parameter in SBA method.

Description

The performance of Stochastic Blockmodel Approximation (SBA) method is contingent on the number of blocks it finds during estimation process, which is rougly determined by a precision parameter delta. cv.SBA tests multiple of delta values to find the optimal one that minimizes the cross validation risk. Note that the optimal delta is not bound to be a single value.

Usage

cv.SBA(A, vecdelta = seq(0.1, 1, by = 0.1))

Arguments

A

either

Case 1.

an (n×n)(n\times n) binary adjacency matrix, or

Case 2.

a vector containing multiple of (n×n)(n\times n) binary adjacency matrices.

vecdelta

a vector containing target delta values to be tested.

Value

a named list containing

optdelta

optimal delta values that minimize the cross validation risk J.

J

cross validation risk values.

References

Chan SH, Airoldi EM (2014). “A Consistent Histogram Estimator for Exchangeable Graph Models.” In Proceedings of the 31st International Conference on International Conference on Machine Learning - Volume 32, ICML'14, I-208–I-216.

Airoldi EM, Costa TB, Chan SH (2013). “Stochastic blockmodel approximation of a graphon: Theory and consistent estimation.” In Burges CJC, Bottou L, Welling M, Ghahramani Z, Weinberger KQ (eds.), Advances in Neural Information Processing Systems 26, 692–700. Curran Associates, Inc.

See Also

est.SBA

Examples

## Not run: 
## generate a graphon of type No.8 with 3 clusters
W = gmodel.preset(3,id=8)

## create a probability matrix for 100 nodes
graphW = gmodel.block(W,n=100)
P = graphW$P

## draw 15 observations from a given probability matrix
A = gmodel.P(P,rep=15)

## cross validate SBA algorithm over different deltas
rescv = cv.SBA(A,vecdelta=c(0.1,0.5,0.9))
print(rescv$optdelta)

## End(Not run)



[Package graphon version 0.3.5 Index]