R devel - Dec 2006 - co.intervals when overlap=0 or 0.5 (PR#9419)

Full_Name: Benjamin Tyner
Version: 2.4.0
OS: linux-gnu
Submission from: (NULL) (71.98.75.54)


As pointed out by Brian Ripley on R-devel, co.intervals does xr <- x[r + ii]
with a fractional r, leading to intervals that don't match the ones returned
by
S-PLUS when overlap=0 or 0.5. For example,

x<-1:10
z<-co.intervals(x,overlap=0) 

As I had already peaked at co.intervals in S-PLUS, I probably shouldn't be
the
one to contribute a fix.

Please don't report bugs that are already fixed in R-devel.

On Sat, 16 Dec 2006, btyner at stat.purdue.edu wrote:
> Full_Name: Benjamin Tyner
> Version: 2.4.0
> OS: linux-gnu
> Submission from: (NULL) (71.98.75.54)
>
>
> As pointed out by Brian Ripley on R-devel, co.intervals does xr <- x[r +
ii]
> with a fractional r, leading to intervals that don't match the ones
returned by
> S-PLUS when overlap=0 or 0.5. For example,
>
> x<-1:10
> z<-co.intervals(x,overlap=0)
>
> As I had already peaked at co.intervals in S-PLUS, I probably shouldn't
be the
> one to contribute a fix.
>
> ______________________________________________
> 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