nnlasso.normal.lambda

beta.old

lambda1

Maximum number of iterations. Default is 10000.

maxiter

xbeta.old

A small value below which a coefficient would be considered as zero. Default is eps=1e-6

Logical. If SE=TRUE, standard errors of the coefficients will be produced. Default is SE=FALSE

The function computes regression coefficients for a penalized generalized linear models subject to non-negativity constraints for a given lambda value for response variable following normal distribution.

internal

Estimates of coefficients of lasso penalized linear regression and generalized linear models subject to non-negativity constraints on the parameters using multiplicative iterative algorithm. Entire regularization path for a sequence of lambda values can be obtained. Functions are available for creating plots of regularization path, cross validation and estimating coefficients at a given lambda value. There is also provision for obtaining standard error of coefficient estimates.

nnlasso.normal.lambda: Coefficients of non-negative penalized generalized linear models for a given lambda for normal family

Description

Usage

Arguments

Value

Details