hublassopath: hublassopath

In Mufabo/Rrobustsp: Robust Signal Processing

hublassopath

Description

hublassopath computes the M-Lasso regularization path (over grid of penalty parameter values) using Huber's loss function.

Usage

hublassopath(y, X, c = NULL, L = 120, eps = 10^-3, intcpt = T,
  reltol = 1e-05, printitn = 0)

Arguments

L	: Positive integer, the number of lambda values EN/Lasso uses. Default is L=120.
reltol	: Convergence threshold for IRWLS. Terminate when successive estimates differ in L2 norm by a rel. amount less than reltol. default: 1e-05
y:	Numeric data vector of size N x 1 (output, respones)
X:	Numeric data matrix of size N x p. Each row represents one observation, and each column represents one predictor (feature) columns are standardized to unit length.
c:	Threshold constant of Huber's loss function (optional; otherwise use default value)
intcpt:	Logical (true/false) flag to indicate if intercept is in the regression mode. Default is true.
eps:	Positive scalar, the ratio of the smallest to the largest Lambda value in the grid. Default is eps = 10^-3.
printitn:	print iteration number (default = 0, no printing)

Value

B : Fitted M-Lasso regression coefficients, a p-by-(L+1) matrix, where p is the number of predictors (columns) in X, and L is the number of Lambda values.

B0 : estimates values of intercepts

stats : structure with following fields:

• Lambda = lambda parameters in ascending order

• sigma = estimates of the scale (a (L+1) x 1 vector)

• gBIC = generalized Bayesian information criterion (gBIC) value for each lambda parameter on the grid.

Mufabo/Rrobustsp documentation built on June 11, 2022, 10:41 p.m.