R package implementing fast statistical algorithms for solving the multivariate normal means problem via empirical Bayes, building on the Extreme Deconvolution method.
Install the package and build the vignette:
remotes::install_github("stephenslab/udr")Note that installing the package will require a C++ compiler setup that is appropriate for the version of R installed on your computer. For details, refer to the CRAN documentation.
Once you have installed the package, load it into your environment:
library(udr)Work through the introductory vignette illustrating the methods on a small simulated data set:
vignette("udr_intro")The udr R package was developed by Yunqi Yang and Peter Carbonetto at the University of Chicago, with contributions from Matthew Stephens.