Choose a News item and you can browse, search, and post messages.
Draft chapters of a book entitled "lme4: Mixed-effects modeling with R" are available under the book link on http://lme4.R-forge.R-project.org/
The three packages, "SASmixed", "mlmRev" and "MEMSS" with examples and data sets for mixed effect models have now been moved from
2009-07-12 (at useR!2009 in Rennes):
Starting with release 0.999375-16, new releases of the lme4 package will be available on CRAN, in addition to the R-forge repository. Much has changed relative to the earlier CRAN versions.
Version 0.999375-1 of the lme4 package is now available on R-forge. This is the alpha test release for lme4_1.0.0. Currently mcmcsamp and nlmer need a bit of work but lmer and glmer should be ok. Please test and report bugs, preferably directly to the bug tracker.
The lme4 package now passes its tests. The mcmcsamp function is still broken so this is not yet the alpha release of lme4_1.0.0
I still haven't debugged the code for determining the conditional modes of the random effects in a GLMM. (It does apparently work for a NLMM.) This is in the C function cond_mode with an interface from R as mer_condMode. I have inserted verbose debugging code triggered by negative values of the verbose argument. For some reason the early iterations appear successful but around iteration 2 or 3 it gets stuck.
The SVN tree for the lme4 package has been moved to R-forge. What is on R-forge now is what had been called the gappy-lmer branch. It is currently under development but usable. The major problem with it at present is that the results for generalized linear mixed models are wrong. The linear mixed models are, as far as I know, working properly and should be faster and more reliable than in earlier versions.
Version 0.99875-8 of the lme4 package has been uploaded to CRAN. This version fixes the problem of missing names of the components of the random effects in the table of estimated variances and correlations of the random effects.