R-Forge Project News

maxLik has moved to GitHub

Arne Henningsen — 2022-05-18T19:08:32+00:00

The development of the maxLik package has been moved to GitHub:

https://r-forge.r-project.org/projects/maxlik/

maxLik has moved to GitHub

Arne Henningsen — 2022-05-18T19:08:25+00:00

The development of the maxLik package has been moved to GitHub:

https://github.com/arne-henningsen/maxLik

maxLik 1.5-2 released

Ott Toomet — 2021-07-30T08:07:39+00:00

Highlights since 1.3-0 include stochastic gradient ascent, tidy/glance methods, and various bugfixes. See NEWS for more information.

maxLik 1.3-0 released

Ott Toomet — 2015-10-28T16:48:07+00:00

This version introduces a new feature, Marquardt (1963) quadratic approximation control for Newton-Raphson related optimizers.

It also introduces a new API: the preferred method to submit the control options to the optimizers is now through 'control=list(...)' argument. The old API is also supported.

Paper about maxLik freely available on-line

Arne Henningsen — 2013-10-21T09:00:52+00:00

An "author-created version" of the paper "maxLik: A package for maximum likelihood estimation in R" (Computational Statistics 26(3), 2011, p. 443-458, written by Arne Henningsen and Ott Toomet) is freely available at Arne's website: https://absalon.itslearning.com/data/ku/103018/publications/maxLik.pdf

maxLik 1.1-2 released

Ott Toomet — 2012-03-04T18:59:58+00:00

This is a bugfix release, improved BHHH handling and more informative error messages for numeric gradient

maxLik 1.0-2 released

Ott Toomet — 2011-10-16T16:23:17+00:00

maxLik 1.0-2 was released today. It is mostly a bugfix release.

* API CHANGE: constrOptim2 removed from API. Names of it's formal arguments are changed. Use maxXXX with the option constraints= instead.

* maxBFGSR works with fixed parameters.

* maxBFGS and other optim-based routines work with both fixed
parameters and inequality constraints.

Article about maxLik in Computational Statistics

Ott Toomet — 2011-10-16T16:21:04+00:00

The article is finally published as Arne Henningsen and Ott Toomet: "maxLik: A package for maximum likelihood estimation in R", Computational Statistics Volume 26, Number 3, 443-458, DOI: 10.1007/s00180-010-0217-1

Available at
http://www.springerlink.com/content/973512476428614p/

Paper about 'maxLik' forthcoming in Comp Stat

Arne Henningsen — 2011-04-06T16:25:57+00:00

A paper that describes the maxLik package will soon be published in the journal "Computational Statistics". It is available on-line at:

http://dx.doi.org/10.1007/s00180-010-0217-1

maxLik 0.7 is available

Arne Henningsen — 2010-04-16T21:55:52+00:00

The "maxLik" package version 0.7 is available on R-Forge and will soon be available in CRAN.

These are the most important changes compared to the previous version (0.6):

* Holding parameters fixed in maxNR() (and hence, also in maxBHHH()) should now be done by the new (optional) argument "fixed", because it is convenient to use than the "old" argument "activePar" in many situations. However, the "old" argument "activePar" is kept for backward-compatibility.

* added (optional) argument "fixed" to functions maxBFGS(), maxNM(), and maxSANN(), which can be used for holding parameters fixed at their starting values

* added function constrOptim2(), which is a modified copy of constrOptim() from the "stats" package, but which includes a bug fix

* added optional argument "cand" to function maxSANN(), which can be used to specify a function for generating a new candidate point (passed to argument "gr" of optim())

* added argument "random.seed" to maxSANN() to ensure replicability

* several mainly smaller improvements in ML estimations with linear equality and inequality constraints (via sumt() and constrOptim2(), respectively)

* several internal changes that make the code easier to maintain