Create Shingle of Conditioning Intervals

USAGE:

equal.count(x, number = 6, overlap = 0.5) 

ARGUMENTS:

x
numeric data.
number
number of intervals.
overlap
controls number of points shared by two successive intervals. If overlap is between 0 and 1, it is the fraction of points shared with its adjacent intervals. If overlap is >=1, it is the number of points shared with adjacent intervals.

VALUE:

object of class "shingle": this is the original data x along with an attribute named intervals which is a matrix with 2 columns and number rows.
The rows of the result, which are used as conditioning intervals, such as in xyplot, describe the endpoints of intervals with the following properties: (1) the first interval has min(x) as its left endpoint and the last interval has max(x) as its right endpoint. (2) All other endpoints are values in x. (2) The numbers of values of x in the intervals are as nearly equal as possible. (4) For two successive intervals, the fraction of points shared by the intervals is as close to overlap as possible.

SEE ALSO:

.

EXAMPLES:

EE <- equal.count(ethanol$E, 9, 0.25) 
xyplot(NOx ~ C | EE, data = ethanol, 
      panel = function(x, y) { 
                panel.xyplot(x, y) 
                panel.loess(x, y, span = 1, degree = 1)})