[R] Superposed histograms

[Frank E Harrell Jr]

On Fri, 10 Jan 2003 12:29:13 +0000 (GMT)
Damon Wischik <djw1005 at cam.ac.uk> wrote:

> 
> >> I woud like to plot cumulative histograms....
> >
> > I don't think this will be effective from a graphical perception point
> > of view.  One problem is that the perception of the bottom symbols will
> > be different than that of the symbols assigned to the upper region,
> > because the upper symbols are not bottom-aligned.
> 
> I _want_ the perception to be different! I gave a toy example, not my real
> application. My real application is this: I have some samples from a
> distribution. Some of the samples I know with high accuracy, some I know
> with low accuracy. I wanted a histogram of the high-accuracy samples drawn
> in a dark colour, and a histogram of the low-accuracy samples plotted on
> top in a different colour. That way my impression would be dominated by
> the part of the plot I am confident about, but I would still see a little
> of the part I am less confident about.
> 
> Do you have any recommendations on how to show graphically a distribution,
> indicating at the same time which parts of the distribution I know with
> confidence?
> 
> Damon Wischik.
> 
> 
My advice would be to plot superposed ECDFs with the one you want to deemphasize shown in light gray scale.
-- 
Frank E Harrell Jr              Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine  http://hesweb1.med.virginia.edu/biostat