Customized(?) VAR

[thwiaethae]

In EViews VAR estimation, if we have two endogenous variables(y1, y2) with lag (1 1),

I know that I have following equations

y1=c(1)+c(2)*y1(-1)+c(3)*y2(-1)+e1
y2=c(4)+c(5)*y1(-1)+c(6)*y2(-1)+e2

but, I wonder that I can estimate following customized(?) equations

y1=c(1)+c(2)*y1(-1)+e1
y2=c(4)+c(5)*y1(+0)+c(6)*y2(-1)+e2

Is that possible estimation in VAR? (My boss told me that this estimation is possible, but I don't know ㅜㅜ)

[EViews Gareth]

No with a VAR. You'll have to use a System instead. You can convert a VAR into a System by using Proc->Make System.