Mixed, low-rank, and sparse multivariate regression (mixedLSR)provides tools for performing mixture regression when the coefficientmatrix is low-rank and sparse. mixedLSR allows subgroup identificationby alternating optimization with simulated annealing to encourage globaloptimum convergence. This method is data-adaptive, automaticallyperforming parameter selection to identify low-rank substructures in thecoefficient matrix.
You can install the development version of mixedLSR fromGitHub with:
# install.packages("devtools")devtools::install_github("alexanderjwhite/mixedLSR")