[Rd] A bug in princomp(), perhaps?

[Ravi Varadhan]

Perhaps, adding Gavin's work around for a dataframe with missing values might also be useful to the documentation:

princomp(na.omit(x))

Thanks,
Ravi

-----Original Message-----
From: r-devel-bounces at r-project.org [mailto:r-devel-bounces at r-project.org] On Behalf Of Ben Bolker
Sent: Monday, June 02, 2014 8:07 PM
To: r-devel at stat.math.ethz.ch
Subject: Re: [Rd] A bug in princomp(), perhaps?

Ben Bolker <bbolker <at> gmail.com> writes:

> 
> FWIW this seems to be a FAQ:
> 
> https://stat.ethz.ch/pipermail/r-devel/2003-July/027018.html
> 
> http://thr3ads.net/r-devel/2013/01/
>    2171832-Re-na.omit-option-in-prcomp-formula-interface-only
> 
> http://r.789695.n4.nabble.com/
>   na-omit-option-in-prcomp-formula-interface-only-td4373533.html
> 
> And two StackOverflow questions (the latter's a bit tangential):
> 
> http://stackoverflow.com/questions/12078291/
>    r-function-prcomp-fails-with-nas-values-even-though-nas-are-allowed
> 
> http://stackoverflow.com/questions/23421438/
>   what-was-wrong-with-running-princomp-in-r/23446938#23446938
> 
>   (Sorry for broken URLs and random assortment of mailing list
> aggregators.)
> 
>   I appreciate Gavin's points that implementing this stuff for 
> princomp.default is difficult/problematic, but I second Ravi's request 
> for a little  more clarification in the help  text; it's quite easy to 
> miss the fact that 'na.action' is defined for princomp.formula but not 
> for princomp.default.  Perhaps just "Note that setting na.action works 
> for princomp.formula, but not for princomp.default" (under the 
> "na.action" argument description).

Putting my effort where my mouth is: I wonder if there is any chance that this patch against the current SVN would be accepted ... ??

Index: princomp.Rd
===================================================================
--- princomp.Rd	(revision 65832)
+++ princomp.Rd	(working copy)
@@ -37,7 +37,8 @@
     when the data contain \code{NA}s.  The default is set by
     the \code{na.action} setting of \code{\link{options}}, and is
     \code{\link{na.fail}} if that is unset. The \sQuote{factory-fresh}
-    default is \code{\link{na.omit}}.}
+    default is \code{\link{na.omit}}.  (This argument applies \emph{only}
+    to the formula method.)}
   \item{x}{a numeric matrix or data frame which provides the data for the
     principal components analysis.}
   \item{cor}{a logical value indicating whether the calculation should
Index: prcomp.Rd
===================================================================
--- prcomp.Rd	(revision 65832)
+++ prcomp.Rd	(working copy)
@@ -40,7 +40,8 @@
     when the data contain \code{NA}s.  The default is set by
     the \code{na.action} setting of \code{\link{options}}, and is
     \code{\link{na.fail}} if that is unset. The \sQuote{factory-fresh}
-    default is \code{\link{na.omit}}.}
+    default is \code{\link{na.omit}}. (This argument applies \emph{only}
+    to the formula method.)}
   \item{\dots}{arguments passed to or from other methods. If \code{x} is
     a formula one might specify \code{scale.} or \code{tol}.}
   \item{x}{a numeric or complex matrix (or data frame) which provides

______________________________________________
R-devel at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel