Package: glamlasso
Type: Package
Title: Penalization in Large Scale Generalized Linear Array Models
Version: 3.0
Date: 2018-01-18
Author: Adam Lund
Maintainer: Adam Lund <adam.lund@math.ku.dk>
Description: 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>.
License: GPL-3
Imports: Rcpp (>= 0.11.2)
LinkingTo: Rcpp, RcppArmadillo
RoxygenNote: 6.0.1
NeedsCompilation: yes
Packaged: 2018-01-19 13:06:30 UTC; adamlund
Repository: CRAN
Date/Publication: 2018-01-19 14:48:29 UTC
