RegrowNetwork: RegrowNetwork

In lorenzha/hdcd: High-Dimensional Changepoint Detection

Description

Prune graph and regrow edges according of scale-free network

Usage

RegrowNetwork(omega, n_nodes = ncol(omega) * 0.1, preferential_power = 1,
  v = 0.3)

Arguments

omega	A precision matrix as created by ScaleNetwork
n_nodes	Number of nodes to prune and regrow. Default is 0.1 of all nodes.
preferential_power	Power coefficient alpha for weighting of degree number as alpha in prefential attachment mechanism.
v	Constant added to the off diagonal of the precision matrix for controlling the magnitude of partial correlations.

Details

Note that v and preferential_power agruments need to be equal to the ones which initially created omega.

Examples

omega <- ScaleNetwork(20, v = 1)
omega_hat <- RegrowNetwork(omega, v = 1)
omega
omega_hat

lorenzha/hdcd documentation built on Sept. 2, 2018, 8:20 p.m.