R help - Jun 2012 - density plot on a log scale

I'm working with a large dataset - large enough that when I do a scatter
plot
the points all blur together, so I want to plot their density by color - a
heat map or something like that.  I've used smoothScatter for tasks like
this, but the problem is that my current dataset really only looks good on a
log-log scale.  When I do the following command

smoothScatter(
        data,
        xlim=c(100,100000000),
        ylim=c(100,10000000000),
        log="xy",
        pch="."
)

the output seems to only show a few of the points, and they are in odd
places that don't seem to correlate at all with the colors.  I got a warning
that I should increase the gridsize, but even using comically large values
(nbin=10000) didn't help.  Does anybody know how I can do this?  Thanks!

--
View this message in context:
http://r.789695.n4.nabble.com/density-plot-on-a-log-scale-tp4633347.html
Sent from the R help mailing list archive at Nabble.com.

On 06/14/2012 07:08 PM, field.cady wrote:
> I'm working with a large dataset - large enough that when I do a
scatter plot
> the points all blur together, so I want to plot their density by color - a
> heat map or something like that.  I've used smoothScatter for tasks
like
> this, but the problem is that my current dataset really only looks good on
a
> log-log scale.  When I do the following command
>
> smoothScatter(
>          data,
>          xlim=c(100,100000000),
>          ylim=c(100,10000000000),
>          log="xy",
>          pch="."
> )
>
> the output seems to only show a few of the points, and they are in odd
> places that don't seem to correlate at all with the colors.  I got a
warning
> that I should increase the gridsize, but even using comically large values
> (nbin=10000) didn't help.  Does anybody know how I can do this? 
Thanks!
>
Hi field.cady,
This is a very rough example (I haven't bothered to do the log 
transformation or anything and the matrix is upside down), but it might 
get you where you want to go.

x<-runif(1000)
y<-runif(1000)
xydens<-matrix(0,10,10)
xybreaks<-seq(0,1,length.out=11)
for(row in 1:10) {
  for(column in 1:10)
   xydens[row,column]<-
   sum(x>=xybreaks[row] & x < xybreaks[row+1] &
   y >= xybreaks[column] & y < xybreaks[column+1])
}
library(plotrix)
color2D.matplot(xydens)

Jim

Would something like ggplot's transparacy option help?
Example bascially from http://had.co.nz/ggplot2/geom_point.html

# Plot large data set.
d <- ggplot(diamonds, aes(carat, price)) + geom_point()
d

# plot large data set with transparancy set to 1/10 th 
p = ggplot(diamonds, aes(carat, price))  + geom_point(alpha = 1/10) 
p







http://had.co.nz/ggplot2/geom_point.html

John Kane
Kingston ON Canada

> -----Original Message-----
> From: jim at bitwrit.com.au
> Sent: Thu, 14 Jun 2012 21:39:36 +1000
> To: field.cady at gmail.com
> Subject: Re: [R] density plot on a log scale
> 
> On 06/14/2012 07:08 PM, field.cady wrote:
>> I'm working with a large dataset - large enough that when I do a
scatter
>> plot
>> the points all blur together, so I want to plot their density by color
-
>> a
>> heat map or something like that.  I've used smoothScatter for tasks
like
>> this, but the problem is that my current dataset really only looks good
>> on a
>> log-log scale.  When I do the following command
>> 
>> smoothScatter(
>>          data,
>>          xlim=c(100,100000000),
>>          ylim=c(100,10000000000),
>>          log="xy",
>>          pch="."
>> )
>> 
>> the output seems to only show a few of the points, and they are in odd
>> places that don't seem to correlate at all with the colors.  I got
a
>> warning
>> that I should increase the gridsize, but even using comically large
>> values
>> (nbin=10000) didn't help.  Does anybody know how I can do this? 
Thanks!
>> 
> Hi field.cady,
> This is a very rough example (I haven't bothered to do the log
> transformation or anything and the matrix is upside down), but it might
> get you where you want to go.
> 
> x<-runif(1000)
> y<-runif(1000)
> xydens<-matrix(0,10,10)
> xybreaks<-seq(0,1,length.out=11)
> for(row in 1:10) {
>   for(column in 1:10)
>    xydens[row,column]<-
>    sum(x>=xybreaks[row] & x < xybreaks[row+1] &
>    y >= xybreaks[column] & y < xybreaks[column+1])
> }
> library(plotrix)
> color2D.matplot(xydens)
> 
> Jim
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
____________________________________________________________
Publish your photos in seconds for FREE
TRY IM TOOLPACK at http://www.imtoolpack.com/default.aspx?rc=if4