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: VaR
See also: VaRcgfDGF2

Quantlet: VaRcharfDG
Description: computes the characteristic function for the class of quadratic forms of Gaussian vectors.

Reference(s):

Link:
Usage: r = VaRcharfDG(t,par)
Input:
t the complex argument to the cgf
par a list defining the distribution; contains at least the following components: theta - the constant term delta - the linear term lambda - the diagonal of the quadratic term
Output:
r the value of the cgf at t

Example:
library("plot")
library("VaR")
library("math")
proc() = VaRcharfDGtest(par,n,xlim)
  dt =(xlim[2]-xlim[1])/(n-1)
  t  = xlim[1] +(0:(n-1))*dt
  r  = VaRcharfDG(complex(t,t*0),par)
  z1 = setmask(t~r.re, "line", "blue")
  z2 = setmask(t~r.im, "line", "red")
  plot(z1,z2)
endp
theta = 0
delta = #(0)
lambda = #(1.4142)
par = list(theta,delta,lambda)
VaRcharfDGtest(par,300,#(-40,40))

Result:
Plots the real (blue line) and the imaginary part (red line)
of the characteristic function for a distribution,
which is close to a chi^2 distribution with one degree of
freedom.



Author: S. Jaschke, 20011026 license MD*Tech
(C) MD*TECH Method and Data Technologies, 05.02.2006