rspa: Adapt numerical records to fit (in)equality restrictions with the Successive Projection Algorithm

Based on (optionally sparse) quadratic optimization with the main algorithms implemented in C. Includes features for easy processing of many (smaller) records. The algorithm has been tested on fairly large optimization problems with up to a few million variables and several hundred thousand restrictions.

Version: 0.1-5
Depends: R (≥ 2.13.0), editrules
Suggests: knitr, testthat
Published: 2014-05-16
Author: Mark van der Loo
Maintainer: Mark van der Loo <mark.vanderloo at>
License: GPL-3
NeedsCompilation: yes
Materials: NEWS
CRAN checks: rspa results


Reference manual: rspa.pdf
Package source: rspa_0.1-5.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: rspa_0.1-5.tgz, r-oldrel: rspa_0.1-5.tgz
OS X Mavericks binaries: r-release: rspa_0.1-5.tgz
Old sources: rspa archive