Page 1 of 1

Impulse response functions

Posted: Wed Apr 28, 2010 12:09 am
by samus_soul
Hi. I having a problem interpreting/understanding the basic impulse response graphs.

I have two I(1) variables in a reduced form VAR model using a 1 s.d shock on their residuals.

When var1 is shocked, the response from var2 graph is it increases by 0.2 s.ds while a shock in var1 on itself causes a response of it to fall by 0.2 s.ds. Does this mean that var1 and var2 have an inverse relationship? I.E the shock/innocation on var1 negatively effects var1 but posititvely effects var2?

Many thanks with any help you can give.

Re: Impulse response functions

Posted: Wed Apr 28, 2010 3:17 pm
by nadja123
Hi, why don't you try to add concrete pictures and tell what your variables are? These should have some economic meaning. Plus try to tell whether you use annual, quarterly or other data, how many lags you included and what factorization you used.

Re: Impulse response functions

Posted: Wed Apr 28, 2010 4:31 pm
by samus_soul
Hi. The variables are logs of fuel exports and manufacturing exports.The data is annual. There is 2 lags to the reduced form VAR (as suggested by AIC criteria).

I have attached the generated responses to this post.

Thanks for any help. :)

Re: Impulse response functions

Posted: Thu Apr 29, 2010 2:03 am
by nadja123
Hi,

1st of all, your impulse responses appear quite unstable. Have you checked out VAR / View / Lag Structure / Lag exclusion test and Lag length criteria? Have you verified the stability of the model under VAR / View / Lag Structure / AR Roots Table?

Interpretation of IRFs: I think they are not correctly estimated, but let me suppose they are correct as they are (I mean sign and evolution...) Then, both your variables yield a positive response (increase) to own positive shock (unexpected increase). The same appears in the case of mutual shocks.

(I do not quite understand why it is interesting to analyze fuel and manufacturing exports in a bivariate VAR.)

Good luck with your work!