CorShrink: Adaptive Shrinkage of Correlation and Covariance Matrices

Performs adaptive shrinkage of correlation and covariance matrices using a mixture model prior over the Fisher z-transformation of the correlations. It allows separate shrinkage intensity for each cell of the correlation matrix. Can also we flexibly extended to shrinking correlation like quantities such as cosine similarities in word2vec models and to partial correlations in conditional graph estimation. For details on methods, refer to Stephens (2017) "False discovery rates: a new deal", <DOI:10.1093/biostatistics/kxw041>.

Package details

Authorc(person("Kushal","Dey",role=c("aut","cre"), email="[email protected]"), person("Matthew","Stephens",role="aut"), person("Peter","Carbonetto",role="ctb"))
MaintainerKushal Dey <[email protected]>
LicenseGPL (>= 2)
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the CorShrink package in your browser

Any scripts or data that you put into this service are public.

CorShrink documentation built on July 13, 2018, 1 a.m.