regularization: Find optimal regularization parameter

In SSPA: General Sample Size and Power Analysis for Microarray and Next-Generation Sequencing Data

Description

Find optimal regularization parameter

Usage

regularization(
  b,
  A,
  beta,
  edf,
  lambda,
  z,
  method = c("lcurve", "gcv", "aic"),
  plot = TRUE,
  log = TRUE,
  verbose = FALSE
)

Arguments

b	the b vector of the system: Ax = b.
A	the A matrix of the system: Ax = b.
beta	regression coefficients.
edf	effective degrees of freedom.
lambda	grid of penalty values.
z	intermediate matrix for L-curvature calculation.
method	Either the L-curve, GCV or AIC.
plot	Plot TRUE/FALSE.
log	Plot on log-scale TRUE/FALSE.
verbose	Verbose TRUE/FALSE

Details

details follow.

Value

generates optionally figure and returns the index for the optimal penalty value.

Author(s)

Maarten van Iterson

SSPA documentation built on Nov. 8, 2020, 5:50 p.m.