EViews.com

Posted: **Thu Oct 02, 2008 12:30 pm**

Does anyone know how to compute the time varying hedge ratio when the estimates are derived from running the GARCH system?

Posted: **Fri Oct 03, 2008 7:01 am**

I'd like to know that as well.

Posted: **Fri Nov 07, 2008 9:50 am**

does anyone know whether there is a way of estimating a multivariate GARCH-M in EViews 6?

Posted: **Thu Nov 13, 2008 10:52 pm**

does anyone know whether there is a way of estimating a multivariate GARCH-M in EViews 6?

EViews 6 doesn't support ARCH-in-mean for multivariate ARCH models. You can roll your own ARCH-in-mean by following the steps below:

1) estimating a GARCH model first without conditional variance in the mean equation
2) generate the conditional variance
3) next create another GARCH system, this time place the conditional variance terms from the first system in the mean equations
4) estimate this system
5) generate the conditional variance with the same name as the first time
6) repeat 4) and 5) until no improvement in model criteria (log likelihood, AIC, etc.)

The above method is a consistent estimator of ARCH-M models.

Here is an example program where y1 and y2 are the two depend variables, but you can do this interactively by hand following the exact same steps:

Code: Select all


' estimate a GARCH model
system sys01
sys01.append y1=c(1)
sys01.append y2=c(2)
sys01.ARCH @DIAGVECH  C(INDEF)   ARCH(1,INDEF)   GARCH(1,INDEF) 
sys01.makegarch

' estimate a GARCH model with garch-in-mean terms
system sys02
sys02.append y1=c(1) + c(3)*garch_01 + c(4)*garch_01_02
sys02.append y2=c(2) + c(3)*garch_01 + c(5)*garch_01_02
sys02.ARCH @DIAGVECH  C(INDEF)   ARCH(1,INDEF)   GARCH(1,INDEF) 
sys02.makegarch

' set convergence criteria
!conv = 0.01

scalar lk = 0

' continue to estimate if difference between successive estimations is greater than !conv
while (abs(sys02.@logl-lk) >!conv)
	sys02.ARCH @DIAGVECH  C(INDEF)   ARCH(1,INDEF)   GARCH(1,INDEF) 
	lk=sys02.@logl 
	sys02.makegarch
wend

Posted: **Thu Dec 25, 2008 6:49 pm**

Excellent tip! thank you! and it is actually much easier in practise than it sounds!

Posted: **Fri Dec 26, 2008 11:59 am**

Excellent tip! thank you! and it is actually much easier in practise than it sounds!

On a second thought this will create the generated regressors problem!!!

EViews.com

GARCH Hedge Ratio Using BEKK

GARCH Hedge Ratio Using BEKK

Re: GARCH Hedge Ratio Using BEKK

multivariate GARCH-M

Re: GARCH Hedge Ratio Using BEKK

Re: GARCH Hedge Ratio Using BEKK