R help - May 2002 - How to shade part of a density plot

I'm trying to shade part of a density plot. The code I'm trying (using
the Old Faithful data as an example) is something like this:

# The Old Faithful geyser data
data(faithful)
d <- density(faithful$eruptions, bw = "sj")
plot(d)
polygon(d[d$x>4], col = "wheat")

I expected that the part of the curve to the right of 4 on the x axis
should be shaded, but nothing gets shaded at all. In fact, when I just
try to print d[d$x>4] I get many lines of 
$"NA"
NULL
I must be mistaken in my understanding of density objects. Could someone
clear up my mistake? Thanks very much.
-- 
Stuart Luppescu       -=- s-luppescu at uchicago.edu
University of Chicago -=- CCSR 
$B:MJ8$HCRF`H~$NIc(B    -=- Kernel 2.4.18-xfs              
checkuary, n:  The thirteenth month of the year.  Begins New
 Year's Day and ends  when a person stops absentmindedly
 writing the old year on his checks. 
 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: This is a digitally signed message part
Url :
https://stat.ethz.ch/pipermail/r-help/attachments/20020519/f5e01aa2/attachment.bin

Stuart Luppescu wrote:
> 
> I'm trying to shade part of a density plot. The code I'm trying
(using
> the Old Faithful data as an example) is something like this:
> 
> # The Old Faithful geyser data
> data(faithful)
> d <- density(faithful$eruptions, bw = "sj")
> plot(d)
> polygon(d[d$x>4], col = "wheat")
> 
> I expected that the part of the curve to the right of 4 on the x axis
> should be shaded, but nothing gets shaded at all. In fact, when I just
> try to print d[d$x>4] I get many lines of
> $"NA"
> NULL
> I must be mistaken in my understanding of density objects. Could someone
> clear up my mistake? Thanks very much.
That's because d is essentially a list, if you do something like

R> sub.x <- d$x[d$x > 4]
R> sub.y <- d$y[d$x > 4]
R> polygon(c(min(sub.x), sub.x), c(0, sub.y), col = "wheat")

you should get the desired result.
Z
> --
> Stuart Luppescu       -=- s-luppescu at uchicago.edu
> University of Chicago -=- CCSR
> $B:MJ8$HCRF`H~$NIc(B    -=- Kernel 2.4.18-xfs
> checkuary, n:  The thirteenth month of the year.  Begins New
>  Year's Day and ends  when a person stops absentmindedly
>  writing the old year on his checks.
> 
> 
>   ------------------------------------------------------------------------
>                        Name: signature.asc
>    signature.asc       Type: application/pgp-signature
>                 Description: This is a digitally signed message part
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Stuart Luppescu wrote:
> 
> I'm trying to shade part of a density plot. The code I'm trying
(using
> the Old Faithful data as an example) is something like this:
> 
> # The Old Faithful geyser data
> data(faithful)
> d <- density(faithful$eruptions, bw = "sj")
> plot(d)
> polygon(d[d$x>4], col = "wheat")
Hmm. Have a look at
  str(d)
to see that d is a density object, which is a list of objects of
different lengths, so indexing as you did doesn't work ...

What about taking 
  index <- d$x > 4
and working with
  d$x[index] and d$y[index]
to generate the shading?

Uwe Ligges

> I expected that the part of the curve to the right of 4 on the x axis
> should be shaded, but nothing gets shaded at all. In fact, when I just
> try to print d[d$x>4] I get many lines of
> $"NA"
> NULL
> I must be mistaken in my understanding of density objects. Could someone
> clear up my mistake? Thanks very much.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Hi all,

I have searched the R-help archives and the bug tracking system and
could not find this one. I am hoping that someone can either reproduce
this or tell me that I am missing a step that is the proximate cause.

I am running R 1.5.0 Patched (May 15, 2002 compiled version kindly
distributed by Prof. Ripley) under WinXP Pro on a P4m platform, which is
up to date with all current patches. The display device is a UXGA 1600 x
1200 LCD panel running in 32 bit mode. If additional HW information is
required, let me know.

The issue appears to be the horizontal positioning of the x-axis labels
when "plot = F" in barplot().

The following simple code segments, if reproducible, should help
demonstrate:
> plot.new()
> midpts <- barplot(1:10, axes = F, axisnames = F)
> axis(1, at = midpts)
The above code results in a barplot with no axis labels initially.  The
call to axis() then results in appropriate x-axis labels under the
midpts of the vertical bars. The x-axis labels are the midpts values.

> plot.new()
> midpts <- barplot(1:10, axes = F, axisnames = F, plot = F)
> axis(1, at = midpts)
The above code, now with "plot = F", results in a barplot with no bars
or axis labels initially.  When I call axis(), the first axis label
("0.7") is offset well to the right, such that it is the only label
displayed.

Am I missing a step that sets the x-axis scaling when "plot = F" or is
this a bug?

If more information is required, let me know.

Thanks,

Marc Schwartz


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._