SQUAREM: Squared Extrapolation Methods for Accelerating EM-Like Monotone Algorithms

Algorithms for accelerating the convergence of slow, monotone sequences from smooth, contraction mappings such as the EM algorithm. It can be used to accelerate any smooth, linearly convergent scheme. A tutorial-style introduction is available in vignette("SQUAREM"). See also: Varadhan & Roland (2008) <doi:10.18637/jss.v092.i07>.

Version:	2026.1
Depends:	R (≥ 4.0)
Suggests:	setRNG, interval
Published:	2026-03-12
DOI:	10.32614/CRAN.package.SQUAREM
Author:	Ravi Varadhan [aut, cre]
Maintainer:	Ravi Varadhan <ravi.varadhan at jhu.edu>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation:	no
Citation:	SQUAREM citation info
Materials:	NEWS
In views:	NumericalMathematics
CRAN checks:	SQUAREM results

Documentation:

Reference manual:	SQUAREM.html , SQUAREM.pdf
Vignettes:	SQUAREM Tutorial (source, R code)

Downloads:

Package source:	SQUAREM_2026.1.tar.gz
Windows binaries:	r-devel: SQUAREM_2026.1.zip, r-release: SQUAREM_2021.1.zip, r-oldrel: SQUAREM_2026.1.zip
macOS binaries:	r-release (arm64): SQUAREM_2026.1.tgz, r-oldrel (arm64): SQUAREM_2026.1.tgz, r-release (x86_64): SQUAREM_2026.1.tgz, r-oldrel (x86_64): SQUAREM_2026.1.tgz
Old sources:	SQUAREM archive

Reverse dependencies:

Reverse depends:	GNE
Reverse imports:	ashr, atena, BayesBrainMap, birdie, codalm, INTACT, INTRIGUE, lava, reReg, spef, templateICAr, tranSurv
Reverse suggests:	BayesfMRI, FixedPoint

Linking:

Please use the canonical form https://CRAN.R-project.org/package=SQUAREM to link to this page.