R help - Apr 2004 - (offtopic) I need two sets of 5 different color scales

Hi,

I am plotting a policy function (result from a dynamic stochastic
optimization problem, discretized approximation).  The policy function
maps from an 2 x 2 x 2 x 3 x B x F state space to a B x F state space
(B and F are usually between 4-6, and represent domestic and foreign
savings.  The other variables are income (Y), inflation (Pi), domestic
and foreign interest rates (R and Z)).  I actually wrote a plotting
function to represent all this, the result is attached -- please have
a look at it and help me...

I need advice in the following: I need two sets of colors for B and F
which are easy to distinguish (when printed on a color laser printer),
represent cardinality (ie have an intuitive mapping to an interval) or
at least ordinality.

I have experimented with the following:

Bcolors <- hsv(.6, seq(0.2, 1, length=5), 1)
Fcolors <- hsv(seq(.1,0, length=5), seq(0.2, 1, length=5)

this is what you see in the plot.  What colors would you use?  Do you
think that varying both brightness and hue helps to distinguish
colors?  Should I change saturation, too?

Thanks,

Tamas

PS.: The plot is simply gzipped.  If you need a zipped version, or the
source code, contact me.

-- 
Tam?s K. Papp
E-mail: tpapp at axelero.hu
Please try to send only (latin-2) plain text, not HTML or other garbage.

On Sat, Apr 10, 2004 at 01:09:18PM +0200, Tamas Papp wrote:
> PS.: The plot is simply gzipped.  If you need a zipped version, or the
> source code, contact me.
It appears to be removed by the list engine.

Is it possible to send eps files to the list?  I gzipped it because it
was 90k originally, but 8k compressed.  What should I do, send the
uncompressed version?

The posting guide says: "No binary attachments except for PS, PDF, and
some image and archive formats", that means no gzipped versions of
these?

Thanks

Tamas

-- 
Tam??s K. Papp
E-mail: tpapp at axelero.hu
Please try to send only (latin-2) plain text, not HTML or other garbage.

Tamas Papp wrote:
> Hi,
> 
> I am plotting a policy function (result from a dynamic stochastic
> optimization problem, discretized approximation).  The policy function
> maps from an 2 x 2 x 2 x 3 x B x F state space to a B x F state space
> (B and F are usually between 4-6, and represent domestic and foreign
> savings.  The other variables are income (Y), inflation (Pi), domestic
> and foreign interest rates (R and Z)).  I actually wrote a plotting
> function to represent all this, the result is attached -- please have
> a look at it and help me...
> 
> I need advice in the following: I need two sets of colors for B and F
> which are easy to distinguish (when printed on a color laser printer),
> represent cardinality (ie have an intuitive mapping to an interval) or
> at least ordinality.
> 
> I have experimented with the following:
> 
> Bcolors <- hsv(.6, seq(0.2, 1, length=5), 1)
> Fcolors <- hsv(seq(.1,0, length=5), seq(0.2, 1, length=5)
> 
> this is what you see in the plot.  What colors would you use?  Do you
> think that varying both brightness and hue helps to distinguish
> colors?  Should I change saturation, too?
Hmm, Ross Ihaka has given a talk at the DSC2003 conference about 
constructing sensible colors. Unfortunately, there is no paper available 
at http://www.ci.tuwien.ac.at/Conferences/DSC-2003/Proceedings/ ...
Some notes and a color package are available at 
http://www.stat.auckland.ac.nz/~ihaka/color/

Also, you might want to look at the package RColorBrewer available at 
CRAN. The package is supposed to provide palettes of sensible colors.

Uwe Ligges


PS: If you want to send any attachments: Preferably upload to a web site 
and just post the link.


> Thanks,
> 
> Tamas
> 
> PS.: The plot is simply gzipped.  If you need a zipped version, or the
> source code, contact me.
> 
> 
> 
> ------------------------------------------------------------------------
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

maybe, the RColorBrewer package does what you want?
see also: ColorBrewer.org
> Hi,
>
> I am plotting a policy function (result from a dynamic stochastic
> optimization problem, discretized approximation).  The policy function
> maps from an 2 x 2 x 2 x 3 x B x F state space to a B x F state space (B
> and F are usually between 4-6, and represent domestic and foreign
> savings.  The other variables are income (Y), inflation (Pi), domestic
> and foreign interest rates (R and Z)).  I actually wrote a plotting
> function to represent all this, the result is attached -- please have a
> look at it and help me...
>
> I need advice in the following: I need two sets of colors for B and F
> which are easy to distinguish (when printed on a color laser printer),
> represent cardinality (ie have an intuitive mapping to an interval) or
> at least ordinality.
>
> I have experimented with the following:
>
> Bcolors <- hsv(.6, seq(0.2, 1, length=5), 1)
> Fcolors <- hsv(seq(.1,0, length=5), seq(0.2, 1, length=5)
>
> this is what you see in the plot.  What colors would you use?  Do you
> think that varying both brightness and hue helps to distinguish
> colors?  Should I change saturation, too?
>
> Thanks,
>
> Tamas
>
> PS.: The plot is simply gzipped.  If you need a zipped version, or the
> source code, contact me.
>
> --
> Tam??s K. Papp
> E-mail: tpapp at axelero.hu
> Please try to send only (latin-2) plain text, not HTML or other garbage.

On Sat, 10 Apr 2004, Tamas Papp wrote:
>
> I need advice in the following: I need two sets of colors for B and F
> which are easy to distinguish (when printed on a color laser printer),
> represent cardinality (ie have an intuitive mapping to an interval) or
> at least ordinality.
It's not offtopic.  This is what the RColorBrewer package is for. If you
look at Cindy Brewer's web version (www.colorbrewer.org) there is
additional information about which scales work well on projectors, cheap color
printers, for people with red-green vision anomalies, and so on.

(It would be nice to have these additional data in the R package)

	-thomas


>
> I have experimented with the following:
>
> Bcolors <- hsv(.6, seq(0.2, 1, length=5), 1)
> Fcolors <- hsv(seq(.1,0, length=5), seq(0.2, 1, length=5)
>
> this is what you see in the plot.  What colors would you use?  Do you
> think that varying both brightness and hue helps to distinguish
> colors?  Should I change saturation, too?
>
> Thanks,
>
> Tamas
>
> PS.: The plot is simply gzipped.  If you need a zipped version, or the
> source code, contact me.
>
> --
> Tam佱s K. Papp
> E-mail: tpapp at axelero.hu
> Please try to send only (latin-2) plain text, not HTML or other garbage.
>
Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle