sparsepca: Sparse Principal Component Analysis (SPCA)

Sparse principal component analysis (SPCA) attempts to find sparse weight vectors (loadings), i.e., a weight vector with only a few 'active' (nonzero) values. This approach provides better interpretability for the principal components in high-dimensional data settings. This is, because the principal components are formed as a linear combination of only a few of the original variables. This package provides efficient routines to compute SPCA. Specifically, a variable projection solver is used to compute the sparse solution. In addition, a fast randomized accelerated SPCA routine and a robust SPCA routine is provided. Robust SPCA allows to capture grossly corrupted entries in the data. The methods are discussed in detail by N. Benjamin Erichson et al. (2018) <arXiv:1804.00341>.

Getting started

Package details

AuthorN. Benjamin Erichson, Peng Zheng, and Sasha Aravkin
MaintainerN. Benjamin Erichson <erichson@uw.edu>
LicenseGPL (>= 3)
Version0.1.2
URL https://github.com/erichson/spca
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:
install.packages("sparsepca")

Try the sparsepca package in your browser

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

sparsepca documentation built on May 2, 2019, 6:37 a.m.