Negative shock

[nazebiah]

I'm trying to do an impulse response with a negative shock.

My model looks like log(x1) log(x2) log(x3) log(x4) log(x5)

But I only want to look at the impulse response of log(x1) and log(x2)

Looking at other posts I believe I need a matrix which I have created.

matrix(2,1) shock
shock.fill(by=c) -1, -1

But when I try to run it I get the error message. "Impulse response factorization must have rows equal to number of endogenous series and columns either 1 or number of endogenous series."

What should my matrix definition be?

Any help would be really appreciated

[nazebiah]

I think I may have worked it out. I've created a new matrix which the correct number of rows to match my model.

matrix(5,1) shock
shock.fill(by=c) 0,-1,0,0,0

I only what to estimate a negative shock from the second variable so I've changed that to a -1 and left the others at 0 is that correct?