XploRe Help : genegarch

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:	times
See also:	genarch

Quantlet:		genegarch
Description:		generates EGARCH(p,q) with Gaussian errors

Reference(s):

Nelson, D.B. (1991) "Conditional Heteroscedasticity in Asset Returns: A new Approach" Econometrica, Vol. 59/2, pp.347-370

Usage:	`{y,s2}=genegarch(alpha,beta,theta,gamma,n)`
Input:
	`alpha`	(q+1)-vector of constant and EGARCH weights
	`beta`	p-vector of GARCH weights
	`theta`	scalar, should be negative for sign effect (Leverage)
	`gamma`	scalar, determines magnitude effect
	`n`	scalar, length of output process
Output:
	`y`	vector, EGARCH process
	`s2`	vector, conditional variance

Note:

The process is sensitive to starting values, so additional care should be taken. This version implements unconditional variance of 0.05.

Example:

library("times") randomize(0) ; - - parameter settings alpha=#(0.1015,0.1342,0.1347,0.078129) beta=#(0.1,0.095) theta=-2 gamma=0.207891 n=1000 ; - - create EGARCH process y=genegarch(alpha,beta,theta,gamma,n) y{1} y{2} graphic=createdisplay(2,1) show(graphic,1,1,setmask(1:n~y{1},"line","thin","blue")) show(graphic,2,1,setmask(1:n~y{2},"line","thin","blue")) setgopt(graphic,1,1,"xlabel", "y", "title", "EGARCH(3,2)","yvalue",0|1) setgopt(graphic,2,1,"xlabel", "y", "title", "conditional Variance","yvalue",0|1)

Result:

graphical output of EGARCH(3,2) process

Author: G. AYDINLI, 20010711 license MD*Tech