EViews.com

Hi all!

I hope this post will help the world! We've been over a full time week on this issue and haven't find a solution yet.

According to the model joined in the message we are trying to fit a bivariate conditional correlation GARCH(1,1) model (Bollerslev, 1990) through a set of data.

The question is, what parameters/variance regressors/ in the system and estimate equation panel need to be selected in order to obtain hi,j,t?

Thanks in advance!

Hunter Simon 2003

Open a new System object

Paste:
WEEKLYBONDRET_US = C(1) + C(2)*WEEKLYBONDRET_US(-1) + C(3)*WEEKLYBONDRET_GER(-1) + C(4)*YCUS(-1) + C(5)*MARKET_SP(-1)

WEEKLYBONDRET_GER = C(6) + C(7)*WEEKLYBONDRET_US(-1) + C(8)*WEEKLYBONDRET_GER(-1) + C(9)*YCGER(-1) + C(10)*MARKET_SP(-1)

Set your GARCH equation

Enjoy

Actually help is much needed for this one! If anyone experienced a CCC GARCH model in Eviews, your toughs are welcome!

The problem is that with the click and play menu of Eviews 6 we can only specify 1 term of error, and this model requires 2 for proper results!

For exemple, if we run the prevous reply code we obtain the following:

Estimation Command:
=====================
ARCH(DERIV=AA, H) @CCC C ARCH(1) GARCH(1)

Estimated Equations:
=====================
WEEKLYBONDRET_US = C(1) + C(2)*WEEKLYBONDRET_US(-1) + C(3)*WEEKLYBONDRET_GER(-1) + C(4)*YCUS(-1) + C(5)*MARKET_SP(-1)

WEEKLYBONDRET_GER = C(6) + C(7)*WEEKLYBONDRET_US(-1) + C(8)*WEEKLYBONDRET_GER(-1) + C(9)*YCGER(-1) + C(10)*MARKET_SP(-1)

Variance and Covariance Representations:
=====================
GARCH(i) = M(i) + A1(i)*RESID(i)(-1)^2 + B1(i)*GARCH(i)(-1)

COV(i,j) = R(i,j)*@SQRT(GARCH(i)*GARCH(j))

BUT, for the Variance and Covariance Representations, we should have:

GARCH(i) = M(i) + A1(i)*RESID(i)(-1)^2 + B1(i)*GARCH(i)(-1) + C1(i)*GARCH(j)(-1)

COV(i,j) = R(i,j)*@SQRT(GARCH(i)*GARCH(j))

Any clues?

Our essay is due for next week, and we have run through all the help available at University. So any help would be appreciated.
We have found many code for tv_garch, but we weren't able to successfully update it according to our model.