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 denxcb denest denci dencb

Quantlet: denxci
Description: computes pointwise confidence intervals with prespecified confidence level for univariate density estimation.

Reference(s):

Usage: {fh, clo, cup} = denxci(x {,h {,alpha {,K} {,xv}}})
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).
alpha confidence level, If not given, 0.05 is used.
K string, kernel function on [-1,1]. If not given, the Quartic kernel "qua" is used.
xv m x 1, values of the independent variable on which to compute the density estimate. If not given, x is used.
Output:
fh n x 2 or m x 2 matrix, the first column is the sorted first column of x or the sorted xv, the second column contains the density estimate on on the values of the first column.
clo n x 2 or m x 2 matrix, the first column is the sorted first column of x or the sorted xv, the second column contains the lower confidence bounds on the values of the first column.
cup n x 2 or m x 2 matrix, the first column is the sorted first column of x or the sorted xv, the second column contains the upper confidence bounds on the values of the first column.

Note:

Example:
library("smoother")
library("plot")
x = 5*normal(200)+10
{fh, clo, cup} = denxci(x,3)
;
fh=setmask(fh,"line","blue")
clo=setmask(clo,"line","blue","thin","dashed")
cup=setmask(cup,"line","blue","thin","dashed")
plot(fh,clo,cup)

Result:
Pointwise confidence intervals at confidence level
alpha = 0.05 for a normal density from N(10,25)
are pictured using Quartic kernel (default) and
bandwidth h=3.



Author: Marlene Mueller 990413
(C) MD*TECH Method and Data Technologies, 05.02.2006