dcgerard/hose: Higher-order spectral estimators

Implements the higher-order spectral estimators (HOSE's) of Gerard and Hoff (2017) <doi:10.1214/17-EJS1330>. Matrix spectral estimators shrink the singular values of a data matrix while keeping the singular vectors unchanged. HOSE's are generalizations of these matrix estimators to tensors. HOSE's usually contain tuning parameters that must be chosen by the user. This package provides methods to calculate these HOSE's and choose these tuning parameters by minimizing an unbiased estimate of the mean squared error. The main functions are get_c(), tensor_var_est(), and soft_coord().

Getting started

Package details

Maintainer
LicenseGPL-3
Version1.0.0
URL http://github.com/dcgerard/hose
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
install.packages("remotes")
remotes::install_github("dcgerard/hose")
dcgerard/hose documentation built on Aug. 1, 2019, 12:11 a.m.