The RATS Software Forum

Hi All,

I am wondering if someone has written a program or procedure on computing Kilian (1998) "Small Sample Confidence Intervals for Impulse Response Functions", RESTATS, confidence bands? Would you be happy to share that with me?

I have written a 2-step principal component estimation of FAVAR and would like to calculate the confidence band.

Thank you.

Shawn

Hey Shawn,
I have a question for you... How this confidence interval different than the one estimated using "MONTESVAR.RPF " ? I assume you want to estimate impulse response functions... I do have relatively small sample, I used "MONTESVAR.RPF " to estimate my impulse functions but I would like to try Kilian's code if it available. Thanks!

sleu123 wrote:Hi All,

I am wondering if someone has written a program or procedure on computing Kilian (1998) "Small Sample Confidence Intervals for Impulse Response Functions", RESTATS, confidence bands? Would you be happy to share that with me?

I have written a 2-step principal component estimation of FAVAR and would like to calculate the confidence band.

Thank you.

Shawn

How would Kilian apply to a model like that? Bootstrapping tends to be very model-specific and I'm not sure how the bias corrections would work for a complicated model like that.

Dear I successfully ran Kilian (1998) using the macromodel file. However, after the line:

spgraph(hfields=3,footer="OLS Responses to unit shock in FFR")
graph(number=0,header="Log IP")
# baseirf(2,7)
graph(number=0,header="Log CPI")
# baseirf(3,7)
graph(number=0,header="GS10")
# baseirf(4,7)
spgraph(done)

for my model it gives error on spgraph and the error is: ## SR6. Missing a Necessary Parameter. Check Instruction Syntax.

Can you please urgently let me know what is wrong in the syntax of spgraph?

It's probably in the GRAPH instruction that follows. Do you have seven shocks in the model? That would only be correct if you have at least seven shocks and FFR is the 7th.