ndeed, I was referring to the methodology proposed by Pesaran and Shin (1998). Well in the absence of this methodology, which you rightly explained, do you think this alternative would work?
proc varmax data=grunfeld plot=impulse;
model y1-y3 = x1 x2 / p=1 lagmax=5
printform=univariate
print=(impulsx=(all) estimates);
I