R help - Nov 2007 - Combining Density plots

Hello,

What I am trying to do is:

Generate a density plot of a population of data.  This data has a bimodal
distribution so I've isolated a couple of possible sub-populations and I
want to overlay these two density plots over the first to see whether they
are contributing to the bimodal population.

I can do this fine with plot(density(...)) and lines(density(...)) .  But
the resulting plots seem to NOT share the same scale or something.  I've had
a read up on bandwidth and fiddled with this but I can't quite get them how
they should look.  Needless to say that the sub-populations are smaller (n)
than the overall population.

Does anyone know how to scale the populations so they all plot on the same
scale with the density function?

Thanks very much

-- 
Gareth Campbell
PhD Candidate
The University of Auckland

P    +649 815 3670
M    +6421 256 3511
E   gareth.campbell@esr.cri.nz
     gcam032@gmail.com

	[[alternative HTML version deleted]]

Hi,
 
you may want to look at Macdonald P (2006) Mixdist package for R
www.math.mcmaster.ca/peter/mix/mix.html
 
It also has an anova test for your model so you would know if your
"possible sub-populations" are statistically significant of not. I
don't know why actually this package is not posted on CRAN, though!
 
I hope this helps,
 
Monica
 
 
-------------------------------------------------
Message: 50Date: Tue, 6 Nov 2007 11:40:07 +1300From: "Gareth Campbell"
<gcam032@gmail.com>Subject: [R] Combining Density plotsTo: "R
Help"
<r-help@r-project.org>Message-ID:<f629fcb80711051440r7d86a04avdea1aab6f9bf135a@mail.gmail.com>Content-Type:
text/plain Hello, What I am trying to do is: Generate a density plot of a
population of data. This data has a bimodaldistribution so I've isolated a
couple of possible sub-populations and Iwant to overlay these two density plots
over the first to see whether theyare contributing to the bimodal population. I
can do this fine with plot(density(...)) and lines(density(...)) . Butthe
resulting plots seem to NOT share the same scale or something. I've hada
read up on bandwidth and fiddled with this but I can't quite get them
howthey should look. Needless to say that the sub-populations are smaller
(n)than the overall population. Does anyone know how to scale the populations so
they all plot on the samescale with the density function? Thanks very much --
Gareth CampbellPhD CandidateThe University of Auckland P +649 815 3670M +6421
256 3511E gareth.campbell@esr.cri.nzgcam032@gmail.com
_________________________________________________________________

Get it now.

6971033
	[[alternative HTML version deleted]]