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: smoother
See also: denxest denci dencb denrot denbwsel denestp

Macro: denest
Description: estimates a univariate density by kernel density estimation. The computation uses WARPing.

Usage: fh = denest(x {,h {,K} {,d} })
Input:
x n x 1 vector, the data.
h scalar, bandwidth. If not given, the rule of thumb bandwidth computed by denrot is used (Silverman's rule of thumb).
K string, kernel function on [-1,1] or Gaussian kernel "gau". If not given, the Quartic kernel "qua" is used.
d scalar, discretization binwidth. d must be smaller than h. If not given, the minimum of h/3 and (max(x)-min(x))/100 is used.
Output:
fh m x 2 matrix, the first column is a grid and the second column contains the density estimate on that grid.

Example:
library("smoother")                                       
library("plot")
;
mu = 10
si = 5
x  = si*normal(200)+mu          ; generate data
;                                  
fh = denest(x)                  ; estimate density
f  = sort(x~pdfn((x-mu)/si)/si) ; true density                   ;             ;                                  
fh = setmask(fh,"line","blue")
f  = setmask(f ,"line","black","thin")
plot(f,fh)                      ; graph functions
Result:
The density estimate (blue) for a normal distribution 
with mean mu=10, standard deviation si=5 is pictured 
using Quartic kernel (default) and Silverman's 
rule-of-thumb bandwidth (default), together with 
the true density (thin black).
Example:
library("smoother")                                       
library("plot")
;
mu = 10
si = 5
x  = si*normal(200)+mu           ; generate data   
;                                  
fhe= denest(x,3,"epa")           ; estimate density
fhu= denest(x,3,"uni")           ; estimate density
f  = sort(x~pdfn((x-mu)/si)/si)  ; true density    
;                              
fhe= setmask(fhe,"line","green")
fhu= setmask(fhu,"line","red")
f  = setmask(f ,"line","black","thin")
plot(f,fhu,fhe)                  ; graph functions
Result:
The density estimate using the Epanechnikov kernel 
(green) is compared to the density estimate using
the Uniform kernel (red) and the true density (thin 
black). In both cases, bandwidth h=3 is used.

Library: smoother
See also: denxest denci dencb denrot denbwsel denestp

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

Author: Wolfgang Haerdle, 910426; Sigbert Klinke, 930219; Lijian Yang, 980108; Marlene Mueller, 990413
(C) MD*TECH Method and Data Technologies, 28.6.1999