ladlassopath: ladlassopath

In Mufabo/Rrobustsp: Robust Signal Processing

ladlassopath

Description

ladlassopath computes the LAD-Lasso regularization path (over grid of penalty parameter values). Uses IRWLS algorithm.

Usage

ladlassopath(y, X, L = 120, eps = 0.001, intcpt = T,
  reltol = 1e-06, printitn = 0)

Arguments

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).
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.
intcpt:	Logical (true/false) flag to indicate if intercept is in the regression model
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 LAD-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. If intercept is in the model, then B is (p+1)-by-(L+1) matrix, with first element the intercept.

stats : structure with following fields: Lambda = lambda parameters in ascending order MeAD = Mean Absolute Deviation (MeAD) of the residuals gBIC = generalized Bayesian information criterion (gBIC) value for each lambda parameter on the grid.

Note

File in Regression.R

Examples

ladlassopath(rnorm(5), matrix(rnorm(5)))

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