mphub: 'phub' with multiple initial points, both 'A' and 'rho' are...

In zhibinghe/Phub: modified EM algorithm for Penalized Hub Model

phub with multiple initial points, both A and rho are randomly drawn from Unif(0,1)

Description

phub with multiple initial points, both A and rho are randomly drawn from Unif(0,1)

Usage

mphub(
  G,
  M,
  lam,
  rep = 10,
  pen.type = c("plog", "plasso", "log"),
  iter.max = 1000,
  tol = 1e-04
)

Arguments

G	observed group data
M	user-specified number of components
lam	tuning parameter for component selection, degenrate to a standard EM without penalty if lam=0
rep	number of repetition
pen.type	type of penalty, including 'log': penalization for logarithm of all components; 'plog': penalization for logarithm of partial components except the null component; 'plasso': penalization for lasso form of partial components
iter.max	maximum iteration steps
tol	threshold for shrinking rho to 0

Value

a list of components

A	M*n matrix containg correlation among nodes
rho	M-dimensional vector containing estimated hub weights
l	log-likelihood
iteration	number of iterations to converge

Examples

n0 = 10; n=100; T=1000
A0 = GenA(100,n0,0.4,0.1,rep(0.05,n))
G0 = GenG(A0,T,c(0.2,rep(0.8/n0,n0)))
M = 15; mphub(G0,M,0.02)$rho
M = 20; mphub(G0,M,0.035)
M = 50; mphub(G0,M,0.019)

zhibinghe/Phub documentation built on Feb. 21, 2025, 11:52 a.m.