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

Library: kernel
See also:

Quantlet: symweigh
Description: computes the symmetrical weights for a selected kernel

Usage: y = symweigh (x, h, n, f)
Input:
x m x 1 vector, n-dimensional origin point
h m x 1 vector or scalar, h[i] contains bandwidth on i-th dimension; if scalar, same bandwidth in all m dimensions
n m x 1 vector or scalar, n[i] contains number of steps on i-th dimension; if scalar, same number of steps in all m dimensions
f string, name of the function computing the selected kernel
Output:
y k x 1 matrix, weights for selected kernel

Example:
library("kernel")
x=(0.1)*(1:4)
h=x
n=10*x+1
y = symweigh(x, h, n, "qua")
y

Result:
Symmetrical weights for selected kernel:

Contents of y
[  1,]  0.0080048
[  2,]  0.007527
[  3,]  0.0061287
[  4,]  0.0057629
[  5,]  0.0035577
[  6,]  0.0033454
[  7,]  0.0039594
[  8,]  0.0037231
[  9,]  0.0030314
[ 10,]  0.0028505
...
[119,]        0
[120,]        0
Example:
library("kernel")
x=(0.1)*(1:4)
h=0.1
n=3
y = symweigh(x, h, n, "qua")
y

Result:
Symmetrical weights for selected kernel:

[ 1,]  0.0013994
[ 2,]  0.0013159
[ 3,]  0.0011824
[ 4,]  0.0012575
[ 5,]  0.0011824
[ 6,]  0.0010624
[ 7,]  0.0010714
[ 8,]  0.0010075
[ 9,]  0.00090527
[10,]  0.0011924
...
[80,]  0.00039727
[81,]  0.00035696



Author: S. Klinke, S. Sperlich, W. Stockmeyer, L. Yang, 19960321 license MD*Tech
(C) MD*TECH Method and Data Technologies, 05.02.2006