Package dpglasso solves the primal formulation of the Graphical Lasso

Description

This package solves the primal formulation of the Graphical Lasso problem, by solving the duals (box QPs) of the block problems.

The outer block-wise optimization routine is written in R with the inner QP solver written in Fortran with a R wrapper around it.

Details

Package: dpglasso
Type: Package
Version: 1.0

There are two functions dpglasso and box_qp_f.

The function dpglasso solves the primal formulation of the Graphical Lasso. The function box_qp_f solves a box constrained Quadratic program via cyclical coordinate descent.

Author(s)

Rahul Mazumder and Trevor Hastie

Maintainer: Rahul Mazumder <rahul.mazumder@gmail.com>

References

The Graphical Lasso: New Insights and Alternatives by Rahul Mazumder and Trevor Hastie, 2011, available at http://arxiv.org/abs/1111.5479

Sparse inverse covariance estimation with the lasso, by J. Friedman , T. Hastie and R. Tibshirani; 2007, Biostatistics.

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.