R help - Mar 2003 - hist() basic question

Hi,
This is a very basic question, but I would like to undestand hist(). I 
thought that the hist( , freq=FALSE) should provide the relative 
frequencies (probabilities), and so they should sum 1, however:

set.seed(2)
ah <- hist(rnorm(100), freq=F)
sum(ah$intensities)
[1] 2

set.seed(2)
bh <- hist(rlnorm(100), freq=F)
sum(bh$intensities)
[1] 0.4999996

I'm getting similar figures with truehist() in MASS.
So I suppose I'm misunderstanding hist(). Any help?

Thanks

Juli

A histogram has area one, not sum one.  From ?truehist

Details:

     This plots a true histogram, a density estimate of total area 1. 


On Sat, 8 Mar 2003, juli g. pausas wrote:
> Hi,
> This is a very basic question, but I would like to undestand hist(). I 
> thought that the hist( , freq=FALSE) should provide the relative 
> frequencies (probabilities), and so they should sum 1, however:
> 
> set.seed(2)
> ah <- hist(rnorm(100), freq=F)
> sum(ah$intensities)
> [1] 2
> 
> set.seed(2)
> bh <- hist(rlnorm(100), freq=F)
> sum(bh$intensities)
> [1] 0.4999996
> 
> I'm getting similar figures with truehist() in MASS.
> So I suppose I'm misunderstanding hist(). Any help?
> 
> Thanks
> 
> Juli
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> 
-- 
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

On Sat, 8 Mar 2003, juli g. pausas wrote:
> Hi,
> This is a very basic question, but I would like to undestand hist(). I
> thought that the hist( , freq=FALSE) should provide the relative
> frequencies (probabilities), and so they should sum 1, however:
No, it provides probability *densities*, which *integrate* to 1.

That is, the height of the bar is the relative frequency divided by the
width of the interval.  This is important because
  - it means histograms with different cutpoints are comparable
  - it means histograms are comparable with mathematical densities such as
a Normal, and with kernel density estimates
  - it means that the bars don't have to have the same width.

If histograms plotted relative frequencies there would be no need to
distinguish them from barplots.

	-thomas