screen_shrink: Shrinkage approach for estimating Gaussian graphical model

In FrankD/nethet: A bioconductor package for high-dimensional exploration of biological network heterogeneity

Description

Shrinkage approach for estimating Gaussian graphical model

Usage

screen_shrink(x, include.mean = NULL, trunc.method = "linear.growth",
  trunc.k = 5)

Arguments

x	The input data. Needs to be a num.samples by dim.samples matrix.
include.mean	Include mean in likelihood. TRUE / FALSE (default).
trunc.method	None / linear.growth (default) / sqrt.growth
trunc.k	truncation constant, number of samples per predictor (default=5)

Value

Returns a list with named elements 'rho.opt', 'wi', 'wi.orig'. Variable rho.opt=NULL (no tuning parameter involved). The variables wi and wi.orig are matrices of size dim.samples by dim.samples containing the truncated and untruncated inverse covariance matrix.

Author(s)

n.stadler

FrankD/nethet documentation built on Oct. 5, 2020, 8:22 a.m.