mOmega: Matrix version solution path function for the Penalized...

View source: R/mOmega.R

mOmegaR Documentation

Matrix version solution path function for the Penalized Interaction Estimation (PIE)

Description

Matrix version solution path function for the Penalized Interaction Estimation (PIE)

Usage

mOmega(Sn, An, lambda = NULL, nlambda = 50, lambda.min.ratio = 0.01,
  err = 10^(-4), maxIter = 10^3, rho = 1)

Arguments

Sn

the sample covariance matrix.

An

the sample kernel matrix.

lambda

user supplied tuning parameter;

err

the precision used to control the convergence. Default is 1e-4.

maxIter

Maximum number of iterations. Default is 1000.

rho

step parameter for the ADMM. Default is 1.

Value

A list of sparse asymmetric p*p matrices, lambda and also the number of iterations.


cescwang85/PIE documentation built on Sept. 10, 2022, 3:28 p.m.