EViews.com

Hi all,

I would be very grateful if someone could help me in this issue:

I want to estimate following equation:

ARCH(THRSH=1, fixedlag=32) BOVA = C(1) + C(2)*VARGJR + C(3) + C(4)*VARGJR*BOVA(-1), where:

BOVA is my dependent variable, the returns of the BOVA 11 ETF traded on the Brazilian Stock exchange and VARGJR is the Variance estimated previously using the GJR Model and incorporated in my model.

The problem is the inclusion of the C(3) term. When I include it, the system acuses "near singular matrix"

I attach the eviews file.

Thanks in advance,

Marcelo

Mathematically impossible.

Sorry, perhaps I did not formulate correctly my question:

I want to formulate following model:

Bova = ALPHA + THETHA* (GARCH) + (GAMMA 1 + GAMMA2*GARCH)* Bova(-1)

GARCH = Estimated as GJR (1,1) and beeing the variance

best Marcelo

P.S> This model is already estimated in the literature