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: stats
See also: stabmom stabcull

Quantlet: stabreg
Description: returns parameter estimates of stable distribution using regression method

Reference(s):

Usage: est=stabreg(x{,method{,epsilon{,maxit}}})
Input:
x n x 1 vector, a sample from a stable distribution
method (optional) string, determines which method should be used, possible options are: "K" - Koutrouvelis method using McCulloch's method as an initial estimation (default), "KM" - Koutrouvelis method using method of moments as an initial estimation, "KW" - Kogon-Williams method.
epsilon (optional) scalar, maximal error, default epsilon = 0.00001, not used in "KW" method
maxit (optional) scalar, maximal number of iterations, default maxit = 5, not used in "KW" method
Output:
est.alpha scalar, estimate of characteristic exponent
est.sigma scalar, estimate of scale parameter
est.beta scalar, estimate of skewness parameter
est.mu scalar, estimate of location parameter

Example:
library("stats")
randomize(100)
x=rndstab(1.3,1,-0.5,0,10000)
est = stabreg(x)
est.alpha
est.sigma
est.beta
est.mu

Result:
Contents of alpha
[1,]   1.2886

Contents of sigma
[1,]   1.0066

Contents of beta
[1,] -0.45274

Contents of mu
[1,]  0.022121
Example:
library("stats")
randomize(100)
x=rndstab(1.7,1,0.5,0,10000)
est = stabreg(x,"K",0.001,6)
est.alpha
est.sigma
est.beta
est.mu

Result:
Contents of alpha
[1,]   1.6978

Contents of sigma
[1,]   1.006

Contents of beta
[1,]  0.47234

Contents of mu
[1,]  -0.022303



Author: S.Borak, R.Weron, 20021023 license MD*Tech
(C) MD*TECH Method and Data Technologies, 05.02.2006