or, using a free piece of software... http://cran.r-project.org/
x <- c(.05,.27,.56,.47,.5,.53,.75,.82,.68,.9) ## data
y <- c(.1,.3,.35,.5,.5,.67,.7,.45,.9,1)
plot(x,y,type="n") ## set up plot area but no points
polygon(x=c(.45,.45,.55,.55),
y=par()$usr[c(3,4,4,3)], ## vertical limits of plot
col=gray(.45), ## desired color
border=F) ## no border
points(x,y) ## now overlay points
[not that this list should become a software help thing, btw ]
-- simon jackman
On May 6, 2006, at 10:28 AM, Moshe Haspel wrote:
> Jamie,
>
> There's probably a better or (certainly more elegant) way to
> accomplish
> this, but overlaying a "range plot with area shading" (rarea)
> should do as
> a quick and dirty solution.
>
> Let's say you had the following 10 data points and wanted to do a
> scatterplot:
>
> IV DV
> .1 .05
> .3 .27
> .35 .56
> .5 .47
> .5 .5
> .67 .53
> .7 .75
> .45 .82
> .9 .68
> 1 .9
>
> 1) create 2 addidional observations to anchor the minimum and maximum
> values of your IV (we started with 10 data points, so we need 12
> all day).
>
> set obs 12
>
> 2) create 2 new variables to form the boundaries of your reference
> area.
>
> gen ref45 = .45
> gen ref55 = .55
>
> 3) set the new observations to hold the minimum and maximum values for
> your IV (For this example, let's assume that the IV has a theoretical
> range of 0 to 1).
>
> replace IV = 0 in 11
> replace IV = 1 in 12
>
> 4) graph! You'll want to customize this output (especially the legend)
> further... but this should get you headed in the right direction.
>
> twoway rarea ref45 ref55 IV, sort bcolor(gs14) || scatter DV IV
>
> NB: STATA does its graphs in the order specified. So if you do scatter
> before rarea, the shading will block out any points within the
> reference
> area. The sort option makes certain that the reference box will
> cover the
> entire area. The bcolor option makes the box in a light gray so
> that the
> points will show up well.
>
> =============================================
> Moshe Haspel, Ph.D.
> Assistant Professor, Department of Political Science
> Spelman College
> Giles 215
> 350 Spelman Lane SW, Box 207
> Atlanta, GA 30314
> 404-270-5654 (phone)
> 404-270-5666 (fax)
>
> Political Methodology Society <[log in to unmask]> wrote on
> 05/05/2006 05:26:17 PM:
>
>> I'm creating a graph in STATA with a yaxis scaled 0 to 1. I want
>> to put
> a
>> shaded region between y = .45 and y = .55. Could someone suggest
>> how to
> do
>> this?
>>
>> Thanks,
>> Jamie Chandler
>> CUNY Grad Center
>>
>> **********************************************************
>> Political Methodology E-Mail List
>> Editor: Karen Long Jusko <[log in to unmask]>
>> **********************************************************
>> Send messages to [log in to unmask]
>> To join the list, cancel your subscription, or modify
>> your subscription settings visit:
>>
>> http://polmeth.wustl.edu/polmeth.php
>>
>> **********************************************************
>
> **********************************************************
> Political Methodology E-Mail List
> Editor: Karen Long Jusko <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
Simon Jackman, Assoc Prof and Director of Graduate Studies,
Dept of Political Science and (by courtesy) Dept of Statistics,
Stanford University, Stanford, CA 94305-6044, USA.
http://jackman.stanford.edu
Director, Political Science Computational Lab. http://pscl.stanford.edu
Program Committee, useR! http://www.r-project.org/useR-2006/
ph: +1 (650) 723-4760 cell: +1 (650) 387 3019
fax: +1 (650) 724-9095
**********************************************************
Political Methodology E-Mail List
Editor: Karen Long Jusko <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|
