XploRe Help : lpregrot

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:	lpregest

Macro:		lpregrot
Description:		determines a rule-of-thumb bandwidth for univariate local polynomial kernel regression using the Quartic kernel.

Reference(s):

Fan and Gijbels (1995): Local Polynomial Fitting

Usage:	`hrot = lpregrot(x {,p})`
Input:
	`x`	n x 2 vector, the data.
	`p`	odd integer <=7, order of polynomial. If not given, p=1 (local linear) is used.
Output:
	`hrot`	scalar, rule-of-thumb bandwidth.

Note:

Internally, the Quartic kernel is used for all computations. If a different kernel function K is given as input, the asymptotical equivalent bandwidth for K is calculated via the canonical kernel approach.

Example:

library("smoother") x=read("nicfoo") hrot=lpregrot(x) hrot

Result:

Contents of hrot
[1,]  0.41215

Library:	smoother
See also:	lpregest

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: Marlene Mueller, 990413