Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Group: Binning, Grids and Sequences
See also: bindata denest grid

Function: conv
Description: conv performs the convolution of a step kernel function and a function over a p-dimensional equidistant grid.

Usage: {xc, yc, fill} = conv (xb, yb, wx, wy {,sym})
Input:
xb m x p x d1 x ... x dn array of integers
yb m x l x d1 x ... x dn array of non-negative integers
wx r x p x d1 x ... x dn array of non-negative integers
wy r x 1x d1 x ... x dn array
sym scalar
Output:
xc k x p x d1 x ... x dn array (k >= m)
yc k x l x d1 x ... x dn array
fill k x 1 x d1 x ... x dn array of zeros or ones

Note:

Example:
x = normal(100)
{xb,yb} = bindata(x, 0.4,0)
wx = #(0,1)
wy = #(1.25,0.75)
{xc, yc, fill} = conv(xb, yb, wx, wy)
cc=createdisplay(1,1)
show(cc,1,1,xc ~ yc)

Result:
Display of a one-dimensional kernel estimate of the normal distribution.
Example:
randomize(0)
x = normal(100,2)
{xb,yb} = bindata(x, #(0.25,0.5),#(0,0))
library("kernel")
x = #(0,0)
h = #(0.25,0.25)
n = #(5,5)
wx= grid(x,#(1,1),n)
wy = qua(wx.*h')
{xc, yc, fill} = conv(xb, yb, wx, wy)
cc=createdisplay(1,1)
show(cc,1,1,xc ~ yc)

Result:
Display of a two-dimensional kernel estimate of the two-dimensional normal distribution.



(C) MD*TECH Method and Data Technologies, 05.02.2006