pzwiernik/golazo: Regularized likelihood for Gaussian graphical models under a general asymmetric lasso-type penalty

The package implements a block-coordinate descent algorithm for the GOLAZO approach to regularized likelihood likelihood for Gaussian graphical model. The penalty is defined by two matrices L, U. The package also allows to perform l1-regularized optimizization without penalizing the diagonal. For details see: Lauritzen, Zwiernik, "Locally associated graphical models", 2020.

install.packages("remotes")
remotes::install_github("pzwiernik/golazo")