Squared Extrapolation Methods for Accelerating EM-Like Monotone Algorithms

Algorithms for accelerating the convergence of slow, monotone sequences from smooth, contraction mapping such as the EM algorithm. It can be used to accelerate any smooth, linearly convergent acceleration scheme. A tutorial style introduction to this package is available in a vignette on the CRAN download page or, when the package is loaded in an R session, with vignette("SQUAREM"). Refer to the J Stat Software article: .


Log of changes to SQUAREM

o first release: August 8, 2010

o second release: August 9, 2010 - fixed a bug in cyclem1' andcyclem2' that croped up when testing under Windows 64-bit. This was due to QR-decomposition using LAPACK.

o third release: September 26, 2010 - modified the `cat' statement such that it is effective only when trace = TRUE

o fourth release: December 31, 2010 - There was a bug in the check for convergence in all the 4 functions: squarem1, squarem2, cyclem1 and cyclem2. It was pointed out by Roger Koenker. It is now fixed. - A slight change was made to documentation regarding the objective function.

o Fifth release: August 10, 2016 - created vignettes with several examples showing how to accelerate convergence using SQUAREM - created demo examples showing how to use SQUAREM to accelerate convergence

o October 19, 2016 - return intermediate parameter vector and merit function values

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


2021.1 by Ravi Varadhan, a year ago


Browse source code at https://github.com/cran/SQUAREM

Authors: Ravi Varadhan

Documentation:   PDF Manual  

Task views: Numerical Mathematics

GPL (>= 2) license

Suggests setRNG

Imported by INTRIGUE, ashr, codalm, lava, reReg, spef, tranSurv.

Depended on by GNE.

Suggested by FixedPoint.

See at CRAN