[R] is zscore() deprecated or Windows only?

Prof Brian Ripley ripley at stats.ox.ac.uk
Sat Aug 30 09:19:20 CEST 2003

Someone has unpacked a Windows binary in one of your library trees, it
seems. The package should refuse to load on MacOS X.

I suggest you search for it and remove it.  If it was in the `full 
install' you installed, complain to the person who made the `install'.
Remember there are two R parts for MacOS X, and I suspect you are using 
the Darwin port (you didn't say) and the mega-bundle made by Jan de Leeuw 
(you didn't say).  If so, there was no such bundle for the current R 
version, 1.7.1, last time I looked.  I suggest you upgrade to 1.7.1, even 
though 1.8.0 is only about 6 weeks' away.

On Sat, 30 Aug 2003, John Christie wrote:

> On Friday, August 29, 2003, at 10:48  PM, Thomas Lumley wrote:
> > On Fri, 29 Aug 2003, Liaw, Andy wrote:
> > This is an unfortunate package name combined with not reading the 
> > output
> > carefully enough.
> >
> > Package: R.base
> > Version: 0.33
> > Date: 2002/10/29
> > Title: [R] Class Library - Stand-alone basic functions
> > Author: Henrik Bengtsson <henrikb at braju.com>
> >
> > So it's in package R.base, which *is* what help.search() says.
> Well, in 1.7.0 full install on Mac OS X (someone made an image with 
> most of the packages and that is what I installed) help.search 
> ("zscore") just returns R.base without any other info (of course, it 
> was the search).  help("zscore") return not found and help("zscore", 
> package=R.base) return "R.base is built for Win32".  Which is very 
> wierd.  

No. Unless you did library(R.base), that is what you should expect.

I tried quitting R (it had been on a week or so) in case some 
> package I loaded removed it or something.  However, that didn't recover 
> it.  Then, I quit R again and lo and behold I get a full help for 
> zscore telling me "Gets the Z scores (standardized residuals)..."  BUT, 
> when I call the function it still says "Error: couldn't find function 
> "zscore".  This is very strange.  

Did you load the library?

> I'll try scale and see if it does 
> what I wanted.  zscore sounded perfect though.  If I could clean up 
> this problem I would, but I don't know if it is a function that was 
> pulled, a function missing, or what the actual state is supposed to be.

Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

More information about the R-help mailing list