[R] plotting circles

[Ross Ihaka]

Here is quick polar scatterplot function (attached) which might give you
some ideas.  This is just a sketch, but shows how to do the basics.
(It contains a number of circles.)

-- 
Ross Ihaka                         Email:  ihaka at stat.auckland.ac.nz
Department of Statistics           Phone:  (64-9) 373-7599 x 5054
University of Auckland             Fax:    (64-9) 373-7018
Private Bag 92019, Auckland
New Zealand
-------------- next part --------------
#  Polar coordinate plot.
#  It is assumed that "r" is given in radians.

polar.plot <-
  function(r, theta,
           pch = NULL,
           col = NULL,
           angle.axis = -90)
  {
    ##  Determine the range of data value.
    ##  Choose pretty cutpoints which include the range.

    rpretty <- pretty(range(abs(r), 0, na.rm=TRUE))
    rmax <- rpretty[length(rpretty)]

    ##  Begin a new plot.
    
    plot.new()

    ##  Set up the plot coordinates.
    ##  We choose a square region which includes a circle
    ##  large enough to include all the data.
    ##  Note the use of asp = 1, so that circles will
    ##  appear as circles, even if a graphics window is resized.
    
    plot.window(xlim = c(-rmax, rmax),
                ylim = c(-rmax, rmax),
                asp = 1)

    ##  Draw a circular grid.
    ##  (a) The circles are really polygons with many vertices.
    ##  (b) The magic number "5" is choosen because a change of
    ##  a change of direction of less than 5 degrees appears smooth.
    
    grid <- seq(0, 2 * pi, length = 360 / 5 + 1)
    for(rad in rpretty) {
      if(rad > 0)
        lines(rad * cos(grid), rad * sin(grid), col = "gray")
    }

    ##  Draw a radial grid in "gray".
    ##  The use of "12" gives divisions into 30 degree slices.
    
    rad <- seq(0, 2 * pi, length = 12 + 1)[-1]
    segments(0, 0, rmax * cos(rad), rmax * sin(rad), col = "gray")

    ##  Basic axis labelling.
    ##  Labels appear along the ray at angle "angle.axis" to the x axis.

    text(rpretty[-1] * cos(angle.axis * pi / 180),
         rpretty[-1] * sin(angle.axis * pi / 180),
         rpretty[-1])
    
    ##  Plot the data values - this is the easy bit.
    
    points(r * cos(theta),
           r * sin(theta),
           col = col, pch = pch)
  }

## Example plot for fake data

group <- sample(3, 100, replace = TRUE)
theta <- 0.5 * rnorm(100) +  0.5 * group
r <- rnorm(100, 4)
polar.plot(r, theta,
           col = c("red","green4","blue")[group],
           pch = 20)