[Lme4-commits] r1781 - in www: . JSS lMMwR misc
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Wed Jun 27 10:39:00 CEST 2012
Author: bbolker
Date: 2012-06-27 10:38:59 +0200 (Wed, 27 Jun 2012)
New Revision: 1781
Modified:
www/JSS/lmer.Rnw
www/index.php
www/lMMwR/ChGLMMBinomial.Rnw
www/lMMwR/ChTheory.Rnw
www/misc/lme4_conversion.md
Log:
small typos in lMMwR
lmer.Rnw: use getME()
index.php: add direct links to SVN, packages, trackers
Modified: www/JSS/lmer.Rnw
===================================================================
--- www/JSS/lmer.Rnw 2012-06-26 08:31:56 UTC (rev 1780)
+++ www/JSS/lmer.Rnw 2012-06-27 08:38:59 UTC (rev 1781)
@@ -1,5 +1,6 @@
\documentclass{jss}
%% need no \usepackage{Sweave.sty}
+\usepackage[american]{babel}
\usepackage{bm,amsmath,thumbpdf,amsfonts}
\author{Douglas Bates\\University of Wisconsin - Madison\And
Martin M\"achler\\ETH Zurich\And
@@ -69,7 +70,7 @@
\setkeys{Gin}{width=\textwidth}
<<preliminaries,echo=FALSE,results=hide>>=
options(width=69,show.signif.stars=FALSE,str=strOptions(strict.width="cut"))
-library(lme4Eigen)
+library(lme4)
@
\begin{document}
\section{Introduction}
@@ -787,15 +788,15 @@
<<densities,fig=TRUE,echo=FALSE,height=5>>=
zeta <- function(m, zmin=-3, zmax=3, npts=301L) {
stopifnot (is(m, "glmerMod"),
- length(m at flist) == 1L, # single grouping factor
+ length(m at flist) == 1L, # single grouping factor
length(m at cnms[[1]]) == 1L) # single column for that grouping factor
- pp <- m at pp
+ pp <- m
rr <- m at resp
- u0 <- pp$u0
- sd <- 1/pp$L()@x
- ff <- as.integer(m at flist[[1]])
- fc <- pp$X %*% pp$beta0 # fixed-effects contribution to linear predictor
- ZL <- t(pp$Lambdat %*% pp$Zt)
+ u0 <- getME(pp,"u")
+ sd <- 1/getME(pp,"L")@x
+ ff <- as.integer(getME(pp,"flist")[[1]])
+ fc <- getME(pp,"X") %*% getME(pp,"beta") # fixed-effects contribution to linear predictor
+ ZL <- t(getME(pp,"Lambdat") %*% getME(pp,"Zt"))
dc <- function(z) { # evaluate the unscaled conditional density on the deviance scale
uu <- u0 + z * sd
rr$updateMu(fc + ZL %*% uu)
Modified: www/index.php
===================================================================
--- www/index.php 2012-06-26 08:31:56 UTC (rev 1780)
+++ www/index.php 2012-06-27 08:38:59 UTC (rev 1781)
@@ -46,7 +46,10 @@
<h2>Resources</h2>
<ul>
<li>A <a href="misc/lme4_conversion.html">guide</a> to changes in the upcoming release of <code>lme4</code> (<a href="misc/lme4_conversion.md">Markdown source</a>, <a href="misc/lme4_compat_report.html">downstream package compatibility report</a>)</li>
-<li>The <a href="http://<?php echo $domain; ?>/projects/<?php echo $group_name; ?>/">project summary page</a>, including links to the source code browser, bug and feature request trackers, etc.</li>
+<li>The <a href="http://<?php echo $domain; ?>/projects/<?php echo $group_name; ?>/">project summary page</a>, including links to the
+<a href="http://r-forge.r-project.org/scm/?group_id=60">source code browser</a>,
+<a href="http://r-forge.r-project.org/tracker/?group_id=60">bug and feature request trackers</a>,
+<a href="http://r-forge.r-project.org/R/?group_id=60">compiled package files</a>, etc.</li>
<li><a href="./doxygen/"><em>Doxygen</em> documentation</a> of the underlying C functions</li>
<li><a href="bib/lme4bib.html">References to articles and other research</a> using <code>nlme</code> or <code>lme4</code>, or the corresponding
<a href="bib/lme4bib.bib">BibTeX file</a>. (If you would like to add your work to this database, please email <code>vasishth.shravan at gmail dot com</code>.)</li>
Modified: www/lMMwR/ChGLMMBinomial.Rnw
===================================================================
--- www/lMMwR/ChGLMMBinomial.Rnw 2012-06-26 08:31:56 UTC (rev 1780)
+++ www/lMMwR/ChGLMMBinomial.Rnw 2012-06-27 08:38:59 UTC (rev 1781)
@@ -12,7 +12,7 @@
library(Matrix)
library(Rcpp)
library(minqa)
-library(lme4a)
+library(lme4)
.f5 = "%.5f"
.f1 = "%.1f"
data(Contraception, package = "mlmRev")
@@ -339,7 +339,7 @@
$(-\infty,\infty)$, and maps it to the probability range, $[0,1]$. It
happens this function is also the cumulative distribution function for
the standard logistic distribution, available in \R{} as the function
-\code{plogit}. In some presentations the relationship between the
+\code{plogis}. In some presentations the relationship between the
logit link and the logistic distribution is emphasized but that often
leads to questions of why we should focus on the logistic
distribution. Also, it is not clear how this approach would
@@ -387,7 +387,7 @@
mixed model but taking into account that slopes or differences in
levels are with respect to the logit or log-odds function. If we wish
to express results in the probability scale then we should apply the
-\code{plogit} function to whatever combination of coefficients is of
+\code{plogis} function to whatever combination of coefficients is of
interest to us.
For example, we see from Fig.~\ref{fig:Contra2} that the observed
Modified: www/lMMwR/ChTheory.Rnw
===================================================================
--- www/lMMwR/ChTheory.Rnw 2012-06-26 08:31:56 UTC (rev 1780)
+++ www/lMMwR/ChTheory.Rnw 2012-06-27 08:38:59 UTC (rev 1781)
@@ -1169,7 +1169,7 @@
Because the link function, $\vec g$, and the inverse link, $\vec
g^{-1}$, are nonlinear functions (there would be no purpose in using a
linear link function) many people use the terms ``generalized linear
-mixed model'' and ``nonlinear mixed model'' interchangably. We
+mixed model'' and ``nonlinear mixed model'' interchangeably. We
reserve the term ``nonlinear mixed model'' for the type of models
used, for example, in pharmacokinetics and pharmacodynamics, where the
conditional distribution is a spherical multivariate Gaussian
Modified: www/misc/lme4_conversion.md
===================================================================
--- www/misc/lme4_conversion.md 2012-06-26 08:31:56 UTC (rev 1780)
+++ www/misc/lme4_conversion.md 2012-06-27 08:38:59 UTC (rev 1781)
@@ -1,13 +1,17 @@
# lme4 release guide
+## To add/FIXME
+* RcppEigen issues for MacOS users
+* mention KR tests etc. in `pbkrtest`?
+
## Version numbering
As [previously announced on the lme4 mailing list][announce], we will shortly be releasing a new version of `lme4`, a descendant of the previous development version `lme4Eigen`. For users who do not access any internal structures, there will be few backward-incompatible changes.
[announce]: https://stat.ethz.ch/pipermail/r-sig-mixed-models/2012q1/014811.html
-* the version of `lme4` currently on R-forge (currently version 0.999902344-0, to be released as version 1.0 (!)) should be used for new projects
-* The current CRAN version (0.999375-42) will be replaced by a nearly identical version called `lme4.0` (currently version 0.9999-2; the only backward-incompatible change in this version is a fix to the AGQ code contributed by Wayne Zhang). `lme4.0` is a maintenance version and will only be changed to fix documented bugs.
+* the version of `lme4` currently on R-forge (currently version 0.999902344-0, to be released as version 1.0 (!)) should be used for new projects.
+* The current CRAN version (0.999375-42) will be replaced by a nearly identical version called `lme4.0` (currently version 0.9999-2; the only backward-incompatible change in this version is a fix to the AGQ code contributed by Wayne Zhang). `lme4.0` is a maintenance version and will only be changed to fix documented bugs. `mer` objects from older versions of `lme4` can be converted to `lme4.0`-usable form via `convert_old_lme4()` in the `lme4.0` package.
* all other versions (`lme4a`, `lme4b`, `lme4Eigen` from R-forge) are deprecated.
## Release notes for end-users
@@ -26,13 +30,14 @@
* `[gn]lmer` now produces objects of class `merMod` rather than class `mer` as before
### New features
-* a general-purpose `getME()` accessor method has been used to allow extraction of a wide variety of components of a mixed-model fit; this has been backported to `lme4.0` for compatibility
+* A general-purpose `getME()` accessor method has been used to allow extraction of a wide variety of components of a mixed-model fit; this has been backported to `lme4.0` for compatibility
* `bootMer`, a framework for obtaining parameter confidence intervals by parametric bootstrapping
* `plot` methods similar to those from the `nlme` package (although missing `augPred`)
-* a `predict` method, allowing a choice of which random effects are included in the prediction
-* likelihood profiling (and profile confidence intervals) for `lmer` and `glmer` results
+* A `predict` method, allowing a choice of which random effects are included in the prediction
+* Likelihood profiling (and profile confidence intervals) for `lmer` and `glmer` results
* `nAGQ=0`, an option to do fast (but inaccurate) fitting of GLMMs
-* negative binomial models
+* Negative binomial models
+* Ability to extract a deviance function for the model, allowing further diagnostics/customization of model results
### Still non-existent features
* Automatic MCMC sampling based on the fit turns out to be very difficult to implement in a way that is really broadly reliable and robust; `mcmcsamp` will not be implemented in the near future
More information about the Lme4-commits
mailing list