secure.control: Internal control function for secure

In secure: Sequential Co-Sparse Factor Regression

Description

list of parameters for controling secure fitting

Usage

secure.control(mu = 1, nu = 1.1, MMerr = 0.001, MMiter = 100,
  outTol = 1e-06, outMaxIter = 200, inMaxIter = 200, inTol = 1e-04,
  lamMaxFac = 1, lamMinFac = 1e-10, gamma0 = 2, elnetAlpha = 0.95,
  spU = 0.25, spV = 0.25)

Arguments

mu	penalty parameter used in enforcing orthogonality
nu	penalty parameter used in enforcing orthogonality (incremental rate of mu)
MMerr	tolerence in the majorization maximization(MM) algorithm for computing initial values when missing value occurs
MMiter	maximum number iterations in the MM algorithm
outTol	tolerence of convergence of outer loop in CURE
outMaxIter	maximum number of outer loop iteration in CURE
inMaxIter	maximum number of inner loop iteration in CURE
inTol	tolerence value required for convergence of inner loop in CURE
lamMaxFac	a multiplier of calculated lambda_max
lamMinFac	a multiplier of determing lambda_min as a fraction of lambda_max
gamma0	power parameter in the adaptive weights
elnetAlpha	elastic net penalty parameter
spU	maximum proportion of nonzero elements in each column of U
spV	maximum proportion of nonzero elements in each column of V

Value

a list of controling parameter.

secure documentation built on Feb. 26, 2021, 5:08 p.m.