[Rd] loess() crashes R on my system

[Prof Brian Ripley]

The posting guide also applies to R-devel: please provide a complete 
reproducible example and don't send HTML, as requested there.

On Fri, 1 Jan 2010, Dennis Murphy wrote:

> Greetings and happy new year!
>
> I am in the process of converting some of the old S-PLUS scripts 
> from Visualizing Data (Cleveland, 1993) into lattice. In fact, I did 
> most of it several years ago, and at the time, all of the scripts 
> that contained loess() worked fine. Tonight, I ran most of the 
> scripts again, but every one that I tried with a loess() call 
> crashed R. I tried it in two sessions, one with an existing .Rdata 
> and in another directory without one. The call below, isolated from 
> its script, is one of the culprits:
>
> with(ethanol, loess(NOx ~ C * E, span = 1/3, degree = 2,
>        parametric = "C", drop.square = "C", family="s"))

What is 'ethanol'?  Is it the data set in package lattice, for 
example?  (If so, this works for me everywhere, including Windows XP.)

> In response to the Posting Guide, when I say 'crash' I mean that the 
> system abnormally terminated R when this line or any other one that 
> invokes loess() is run. I'm running the binary version of 2.10.1 on 
> a Dell Studio 15 with Windows 7 Home Premium.

If you mean any call to loess(), the problem is your machine (or at 
least your precise version of the OS) since example(loess) has been 
run on the binary version of R before distribution.  The examples are 
always a good place to start testing.)

Problems specific to the Windows binary (and we don't know that as yet 
but it seems likely) should be reported to the R-windows address.

> This is the first time I've ever experienced something like this in 
> the decade that I've used R. What's more surprising is that it 
> occurred with such a common function. Since I haven't seen any 
> postings of this type, it's probably something on my end, but I 
> don't know what. Since it shut down the program, I felt compelled to 
> report it.

You should be able to get a problem report under those circumstances, 
usually as an option on the dialog box -- and you will need to report 
that with the reproducible example.

> I am subscribed to r-help but not r-devel, so please cc any response to me.
> Thanks in advance for
> any assistance.
>
> Dennis Murphy
>
>> sessionInfo()
> R version 2.10.1 (2009-12-14)
> i386-pc-mingw32
>
> locale:
> [1] LC_COLLATE=English_United States.1252
> [2] LC_CTYPE=English_United States.1252
> [3] LC_MONETARY=English_United States.1252
> [4] LC_NUMERIC=C
> [5] LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> loaded via a namespace (and not attached):
> [1] tools_2.10.1
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

-- 
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