## Significance of impulse response VAR

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

sofp
Posts: 13
Joined: Mon Jul 04, 2016 2:48 pm

### Significance of impulse response VAR

Hi,

i'm estimating an unrestricted VAR and right now I went thought the impulse response function. I don't understand how to read the significance of the IRF. For example I don't understand on which basis from the IRF graph I can say that real GDP response to oil price shock is whether significant or insignificant.

Thanks.

dakila
Posts: 452
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Significance of impulse response VAR

If the confidence interval (or band) does not contain zero (horizontal axis) then it is statistically significant.

sofp
Posts: 13
Joined: Mon Jul 04, 2016 2:48 pm

### Re: Significance of impulse response VAR

Thank you for your answer. I'm still a little bit confused about it. So can I ask you to give me an example based one these impulse responses? Thanks
Attachments
Schermata 2016-08-07 alle 10.11.33.png (24.79 KiB) Viewed 18163 times

dakila
Posts: 452
Joined: Tue Nov 24, 2015 4:57 pm

### Re: Significance of impulse response VAR

dlogrgdp to dlogop: impulse responses are insignificant for all periods
dlogop to dlogop: impulse responses are significant for 1st period (impact period)
dlogstir to dlogop: impulse responses are significant for 2nd period
dlogltir to dlogop: impulse responses are significant for 1st period (impact period)

sofp
Posts: 13
Joined: Mon Jul 04, 2016 2:48 pm

### Re: Significance of impulse response VAR

Thank you so much!

soetirl13
Posts: 2
Joined: Wed May 16, 2018 10:24 am

### Re: Significance of impulse response VAR

Hi,

just wondering. What does it mean when it's insignificant? is it meaningless then??

startz
Non-normality and collinearity are NOT problems!
Posts: 3747
Joined: Wed Sep 17, 2008 2:25 pm

### Re: Significance of impulse response VAR

All it means is that there is not strong statistical evidence that the response is different from zero.

ericsugandi
Posts: 3
Joined: Thu Mar 18, 2021 10:07 pm

### Re: Significance of impulse response VAR

@dakila: Is there any textbook or other material that I can refer to on how to interpret IRF confidence intervals as you did? Thanks very much for your help.

virenrehal
Posts: 3
Joined: Thu Sep 22, 2022 3:07 am

### Re: Significance of impulse response VAR

As requested, this link contains information on the estimation and interpretation of IRFs (Impulse Responses) for both VAR and VECM: https://spureconomics.com/impulse-response-functions-after-var-and-vecm/

The interpretation is quite straightforward, although it does depend on how the IRFs are computed. Generally, they are computed using shocks in terms of Standard Deviation or in terms of absolute values of the variables.

Posts: 8
Joined: Thu Dec 24, 2020 1:49 am

### Re: Significance of impulse response VAR

Hello everyone
I have a question on the same matter
What is the link between causality, PVAR findings, and impulse response (IRF) in PVAR? If the causality shows a significant bidirectional relationship between X and Y variables, do the IRF should reflect the same? Also, if the PVAR results show X has a significant impact on Y, do the IRF should reflect the same?
Feedback is highly appreciated