snowboot (version 0.5.1)

bootdeg: Bootstrapping Empirical Degree Distribution

Description

This function delivers a bootstrap estimate of network degree distribution based on a LSMI sample. Default is one bootstrap replication.

Usage

bootdeg(sam.out, num.sam = sam.out$num.sam, n.boot = 1)

Arguments

sam.out

A list that is the output of Oempdegreedistrib.

num.sam

A vector of integers containing the numeric ids of the LSMI samples when sam.out$num.sam is greater than one. When num.sam is an integer, N, LSMI from 1 to N are taken from the input sam.out.

n.boot

A positive integer number, the number of bootstrap replications.

Value

A list consisting of:

values

A list of length num.sam where each element is a vector containing the unique degree values sampled in each LSMI.

empd

A list of length three where each element contains a different estimate of degree distribution: empd.w.p0s - weighted bootstrap with a proportion of isolated nodes p0 being estimated by simple random sampling of bootstrapped seeds; empd.nw.p0sEkb - non-weighted bootstrap with a proportion of isolated nodes p0 being estimated by simple random sampling of bootstrapped seeds; empd.nw.p0sEks -ignore (see Thompson et al. for details)

num.sam

Numeric indices corresponding to LSMI samples used for bootstrap.

n.boot

The same object as input argument n.boot.

n.neigh

The number of waves carried out by the snowball sample. This is the same value from sam.out$n.neigh.

seeds1

A matrix of dimension lenght(num.sam) x n.seeds with the numeric seed ids. Each row corresponds to one LSMI. The rows are present in the same order as the ids in num.sam.

nodes_of_LSMI

A list of length length(num.sam) where each element is vector containing the numeric ids of the nodes sampled using the respective LSMI. The elements are present in the same order as the ids in num.sam. Note: nodes_of_LSMI is unreported when n.neigh equals zero.

References

Efron, B. (1979). Bootstrap methods: another look at the jackknife. The annals of Statistics, 1-26.

Thompson, M. E., Ramirez Ramirez, L. L., Lyubchich, V. and Gel, Y. R. (2015), Using the bootstrap for statistical inference on random graphs. Can J Statistics. doi: 10.1002/cjs.11271

Examples

net <- artificial_networks[[1]]
sam.out <- Oempdegreedistrib(net = net, n.seeds = 40, n.neigh = 1, num.sam = 1)
a <- bootdeg(sam.out = sam.out, n.boot = 50)