SVAR with cointegration restrictions a la Warne (1993)

[pfb382]

Hi,

I have a quite specific question, but lets see. Would appreciate help very much!
I am running a cointegration model, which I identify based on Anders Warne(1993)'s common trend model.
http://www.texlips.net/awarne/code.html

In the webpage above he provides the code, where file ctair.src contains the calculations for the impulse response functions on lines 151-160:

do I1=1,Nper+1
do i=1,n
do j=1,n
comp RjSer(i,j)(I1) = Rjmat(I1)(i,j)
comp SRjSer(i,j)(I1) = SRjmat(I1)(i,j)
comp RjseSer(i,j)(I1) = Rjsemat(I1)(i,j)
comp SRjseSer(i,j)(I1) = SRjsemat(I1)(i,j)
end do j
end do i
end do I1

Rjmat is for the first diff responses and SRjmat is for the levels responses, an accumulation of Rjmat. The Rjsemat is the asymptotic standard error of the Rjmat estimate.

Right now the code gives impulse responses for 1 std deviation. But as I need to compare similar shocks across different models/countries, I need 1% shocks instead.
I have been in contact with Warne, he told me I need to rescale the right hand side of the equality signs in the comp commands. But I dont know how this is done. Do you have an idea?
I attached the entire Ctair.src file here as well.

Would be grateful for any hints.
Thanks!