Help - Writing a VEC Model
Posted: Fri Mar 23, 2012 4:18 am
Hello. I'm new on this forum and i'm still getting used with EViews. I need some support with writing VEC models. In the attached file, I have 4 series: BRENT, DBRENT (which is the first order difference of BRENT), SOJA and DSOJA (first order difference of SOJA). I'm willing to write a VEC(1) model with this 4 series and then estimate with VECH-GARCH,BEKK-GARCH etc. However, I don't know for sure how to write this model. I've tried several times and I always get the same message from EViews ("Near Singular Matrix"), suggesting the existence of perfect multicollinearity on my data.
I'm trying to write something like this:
DSOJA = C(1)*(SOJA(-1)+C(2)*BRENT(-1))+C(3)*DSOJA(-1)+C(4)*DBRENT(-1)+C(5)
DBRENT = C(6)*(SOJA(-1)+C(2)*BRENT(-1))+C(7)*DSOJA(-1)+C(8)*DBRENT(-1)+C(9)
I don't know for sure whether I should include equations for SOJA and BRENT too, since I'm working with a system of 4 variables...
Please, I need some insight from you guys. Any VEC model that works... I can't just use SOJA and BRENT and use estimate VAR(VEC) on EViews since they are not I(0) and if use DBRENT and DSOJA, EViews would write a VEC model with D(DBRENT) and D(DSOJA), which would be the second differences of BRENT and SOJA.
Thanks in advance!
I'm trying to write something like this:
DSOJA = C(1)*(SOJA(-1)+C(2)*BRENT(-1))+C(3)*DSOJA(-1)+C(4)*DBRENT(-1)+C(5)
DBRENT = C(6)*(SOJA(-1)+C(2)*BRENT(-1))+C(7)*DSOJA(-1)+C(8)*DBRENT(-1)+C(9)
I don't know for sure whether I should include equations for SOJA and BRENT too, since I'm working with a system of 4 variables...
Please, I need some insight from you guys. Any VEC model that works... I can't just use SOJA and BRENT and use estimate VAR(VEC) on EViews since they are not I(0) and if use DBRENT and DSOJA, EViews would write a VEC model with D(DBRENT) and D(DSOJA), which would be the second differences of BRENT and SOJA.
Thanks in advance!