Penalized Precision Matrix Estimation via ADMM

Estimates a penalized precision matrix via the alternating direction method of multipliers (ADMM) algorithm. It currently supports a general elastic-net penalty that allows for both ridge and lasso-type penalties as special cases. This package is an alternative to the 'glasso' package. See Boyd et al (2010) for details regarding the estimation method.


News

ADMMsigma 2.1

  • Lots and lots of minor bug fixes

  • Added several (some modified) vignettes for clarity on the underlying algorithm and utility of the package (see the package website)

  • Reversed the ADMM algorithm sequence -- which improves computation speed

ADMMsigma 2.0

  • ADMMsigma is a major version release and total re-work of the original package.

  • Updates include further parallel computing capabilities, automatic selection of tuning parameter grid, and updated graphics. Many options for "ADMMsigma" have been added and/or re-named. Please refer to the manual for more detailed information.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("ADMMsigma")

2.1 by Matt Galloway, a year ago


https://github.com/MGallow/ADMMsigma


Report a bug at https://github.com/MGallow/ADMMsigma/issues


Browse source code at https://github.com/cran/ADMMsigma


Authors: Matt Galloway [aut, cre]


Documentation:   PDF Manual  


GPL (>= 2) license


Imports stats, parallel, foreach, ggplot2, dplyr

Depends on Rcpp, RcppProgress, doParallel

Suggests testthat, knitr, rmarkdown, microbenchmark, pkgdown

Linking to Rcpp, RcppArmadillo, RcppProgress

System requirements: GNU make


See at CRAN