[R] problems with hist()

Prof Brian Ripley ripley at stats.ox.ac.uk
Tue May 20 17:55:01 CEST 2003


As a last resort, try read the help page:

     A numerical tolerance of 1e-7 times the range of the breaks is
     applied when counting entries on the edges of bins. 

If you use a silly range, this will affect your results.


On Tue, 20 May 2003, Guenter Tirler wrote:

> Hi,
> I try to compute the counts of a histogram of random numbers. But I
> get different results depending on the limits (-10,10) or (-1000,1000)
> although all numbers are in the interval (-10,10).
> 
> Have I done somthing wrong? 
> 
> 
> > x<-rnorm(1e6)
> > s1<-hist(x, br =c(-1000,-1,-0.5,0,0.5,1,1000),plot=FALSE)[[2]]
> > s2<-hist(x, br =c(-10,-1,-0.5,0,0.5,1,10),plot=FALSE)[[2]]
> > sum(s1)
> [1] 1000000
> > sum(s2)
> [1] 1000000
> >  s1                                                        
> [1] 159035 149461 192003 191478 149789 158234
> > s2
> [1] 159009 149452 192004 191464 149805 158266

-- 
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