[R] Cumulative vs. non-cumulative IRFs in R

mrrox r.otojanov at qmul.ac.uk
Sat Aug 15 18:04:58 CEST 2015


I am using irf function from vars package. I am trying to derive cumulative
IRFs.

The following code describes the case of deriving cumulative IRFs:

irf(vecm.l, impulse = c("g","p","h","l","s"), response = "g",  cumulative =
TRUE,n.ahead = 20, ortho=TRUE)

I got the output and plotted it, it looked like cumulative values of
estimated MA coefficients. But then I ran the code with cumulative switched
to FALSE as below:

irf(vecm.l, impulse = c("g","p","h","l","s"), response = "g",  cumulative =
FALSE,n.ahead = 20, ortho=TRUE)

Output from these two codes is identical. Including cumulative = TRUEin the
irf function does not produce the cumulative responses??

Please note that I converted a vecm model to var using vec2var, hence the
input model is called vecm.l here. Though, I doubt it plays any significance
in deriving cumulative IRFs.

I would appreciate your comment, am I specifying the function incorrectly?
Thanks,



--
View this message in context: http://r.789695.n4.nabble.com/Cumulative-vs-non-cumulative-IRFs-in-R-tp4711138.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list