sure_rank: Iterates through all multilinear ranks less than 'max_nrank'...

In dcgerard/hose: Higher-order spectral estimators

Description

Iterate through all ranks less than max_nrank and choose rank with smallest sure.

Usage

sure_rank(X, max_nrank = dim(X), tau2 = 1)

Arguments

X	An array of numerics. The data.
max_nrank	A vector of positive integers. The maximum rank to inspect for each mode.
tau2	The variance, assumed known. Defaults to 1.

Value

min_rank The multilinear rank that minimizes the SURE.

min_sure The minimum SURE value.

sure_vec A vector of all the SURE values of all the multilinear ranks that were inspected.

all_ranks A matrix of all of the multilinear ranks that were inspected.

est The final mean estimate.

Author(s)

David Gerard.

References

Gerard, D., & Hoff, P. (2015). Adaptive Higher-order Spectral Estimators. arXiv preprint arXiv:1505.02114.

dcgerard/hose documentation built on Aug. 1, 2019, 12:11 a.m.