| Library: | smoother |
| See also: | denbwcrit |
| Quantlet: | gauder | |
| Description: | gauder evaluates derivatives of the Gaussian kernel rescaled by a bandwidth h, to be used for density estimation bandwidth selection. |
| Usage: | y = gauder (x, h, p) | |
| Input: | ||
| x | n x 1 vector | |
| h | scalar, bandwidth | |
| p | scalar, order of derivative | |
| Output: | ||
| y | n x 1 matrix, Gaussian density derivatives | |
library("smoother")
x = aseq(-1, 21, 0.1)
y = gauder(x,1,3)
t = createdisplay(1,1)
show(t,1,1,x~y)
plot the third normal derivative, bandwidth = 1