benjaminfrot/lrpsadmm: Low-Rank Plus Sparse Estimation via Alternating Direction Method of Multipliers (ADMM)

Fit the Low-Rank plus Sparse (LRpS) estimator using the Alternating Direction Method of Multipliers (ADMM). This model learns an inverse covariance matrix which is the sum of a sparse matrix and a low-rank matrix as suggested by Chandrasekaran et al (2012) (DOI:10.1214/11-AOS949). The package supports robust estimation via an estimator of the correlation matrix based on Kendall rank correlations. It includes functions to compute a whole regularisation path and to select tuning parameters with cross-validation.

install.packages("remotes")
remotes::install_github("benjaminfrot/lrpsadmm")