glamlasso: Penalization in Large Scale Generalized Linear Array Models

Functions capable of performing efficient design matrix free penalized estimation in large scale 2 and 3-dimensional generalized linear array model framework. The generic glamlasso() function solves the penalized maximum likelihood estimation (PMLE) problem in a pure generalized linear array model (GLAM) as well as in a GLAM containing a non-tensor component. Currently Lasso or Smoothly Clipped Absolute Deviation (SCAD) penalized estimation is possible for the followings models: The Gaussian model with identity link, the Binomial model with logit link, the Poisson model with log link and the Gamma model with log link. Furthermore this package also contains two functions that can be used to fit special cases of GLAMs, see glamlassoRR() and glamlassoS(). The procedure underlying these functions is based on the gdpg algorithm from Lund et al. (2017) <doi:10.1080/10618600.2017.1279548>.

Getting started

Package details

AuthorAdam Lund
MaintainerAdam Lund <[email protected]>
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the glamlasso package in your browser

Any scripts or data that you put into this service are public.

glamlasso documentation built on May 2, 2019, 2:18 a.m.