Convex banding of the covariance matrix using


hierband is the R package implementing the convex banding approach to covariance estimation of Bien, Bunea, & Xiao (see full reference below).


The package is designed for situations in which there is a large number of variables that have a known ordering and in which it is believed that variables far apart in this ordering have little to no dependence.

It is called hierband (pronounced "hair band") because it makes use of a hierarchical group lasso penalty and provides a banded estimate of the covariance matrix.

Its main functions are hierband, hierband.path,


Jacob Bien, Florentina Bunea, Luo Xiao


Bien, J., Bunea, F. Xiao, L. (2014) "Convex banding of the covariance matrix." Accepted for publication in JASA.

Questions? Problems? Suggestions? or email at

Please suggest features or report bugs with the GitHub issue tracker.

All documentation is copyright its authors; we didn't write any of that.