[Lme4-commits] r1797 - www/JSS

[noreply at r-forge.r-project.org]

Author: bbolker
Date: 2013-02-14 05:19:46 +0100 (Thu, 14 Feb 2013)
New Revision: 1797

Modified:
   www/JSS/glmer.Rnw
   www/JSS/lmer.Rnw
Log:
more JSS updates



Modified: www/JSS/glmer.Rnw
===================================================================
--- www/JSS/glmer.Rnw	2013-02-14 00:19:41 UTC (rev 1796)
+++ www/JSS/glmer.Rnw	2013-02-14 04:19:46 UTC (rev 1797)
@@ -42,6 +42,13 @@
   1280 Main Street W \\
   Hamilton, ON L8S 4K1, Canada \\
   E-mail: \email{bolker at mcmaster.ca}
+  \par\bigskip
+  Steven C. Walker\\
+  Department of Mathematics \& Statistics \\
+  McMaster University \\
+  1280 Main Street W \\
+  Hamilton, ON L8S 4K1, Canada \\
+  E-mail: \email{scwalker at math.mcmaster.ca }
 }
 \newcommand{\Var}{\operatorname{Var}}
 \newcommand{\abs}{\operatorname{abs}}

Modified: www/JSS/lmer.Rnw
===================================================================
--- www/JSS/lmer.Rnw	2013-02-14 00:19:41 UTC (rev 1796)
+++ www/JSS/lmer.Rnw	2013-02-14 04:19:46 UTC (rev 1797)
@@ -167,15 +167,18 @@
     \textbf{Symbol} & \textbf{Meaning} & \textbf{lme4 equivalent} \\
     \hline
     $\bm X$ & Fixed-effect design matrix & \code{getME(.,"X")} \\
-    $\bm Z$ & Sparse random-effect design matrix & \code{t(getME(.,"Zt")} \\
+    $\bm Z$ & Sparse random-effect design matrix & \code{getME(.,"Z")} \\
     $\bm \theta$ & Variance-covariance parameter vector (log-Cholesky scale) &
     \code{getME(.,"theta")} \\
-    $\bm \beta$ & Fixed-effect coefficients & \code{fixef(.)} \\
-    $\sigma^2$ & Residual variance & \\
+    $\bm \beta$ & Fixed-effect coefficients & \code{fixef(.)}  [\code{getME(.,"beta")}]\\
+    $\sigma^2$ & Residual variance & \verb+sigma(.)^2+ \\
     $\mc{Y}$ & Response variable & \\
     $\bm\Lambda_{\bm\theta}$ & Relative covariance factor & \\
+    $\bm L_\theta$ & Sparse Cholesky factor & \code{getME(.,"Lambda")} \\
     $\mc B$ & Random effects & \\
     $\mc U$ & Spherical random effects & \\
+    $\bm u$ & Conditional modes of random effects & \code{ranef(.)} [\code{getME(.,"u")}] ? \\
+    $\bm P$ & Fill-reducing permutation & 
   \end{tabular}
   \caption[Table of notation]{Table of notation and \pkg{lme4} equivalents
   \bmb{finish this table!}
@@ -577,9 +580,15 @@
 predictions?
 
 \section{Discussion}
-Is there anything else to say here?  Point to alternatives
-in the FOSS ecosystem?  Open problems?
+What do we want to say here?
 
+\begin{itemize}
+\item Point to alternatives in the R/FOSS ecosystem
+\item State $p$-value issues for the record
+\item \emph{Advantages} of \pkg{lme4}: fast, handles multiple/crossed RE, 
+  provides profiling/parametric bootstrapping; flexible (sort of)
+\item \emph{Disadvantages}: no R-side effects; (no $p$-values); not as flexible as e.g. AS-REML
+\end{itemize}
 
 \bibliography{lmer}
 \end{document}