FRESHD: Fast Robust Estimation of Signals in Heterogeneous Data

Procedure for solving the maximin problem for identical design across heterogeneous data groups. Particularly efficient when the design matrix is either orthogonal or has tensor structure. Orthogonal wavelets can be specified for 1d, 2d or 3d data simply by name. For tensor structured design the tensor components (two or three) may be supplied. The package also provides an efficient implementation of the generic magging estimator.

Getting started

Package details

AuthorAdam Lund [aut, cre, ctb, cph], Benjamin Stephens [ctb, cph], Gael Guennebaud [ctb, cph], Angelo Furfaro [ctb, cph], Luca Di Gaspero [ctb, cph], Brandon Whitcher [ctb, cph]
MaintainerAdam Lund <adam.lund@math.ku.dk>
LicenseGPL
Version1.0
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:
install.packages("FRESHD")

Try the FRESHD package in your browser

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

FRESHD documentation built on May 12, 2022, 9:06 a.m.