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

Postby sofp » Fri Aug 05, 2016 3:02 pm

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: 334
Joined: Tue Nov 24, 2015 4:57 pm

Re: Significance of impulse response VAR

Postby dakila » Fri Aug 05, 2016 6:28 pm

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

Postby sofp » Sun Aug 07, 2016 2:14 am

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
Schermata 2016-08-07 alle 10.11.33.png (24.79 KiB) Viewed 5232 times

dakila
Posts: 334
Joined: Tue Nov 24, 2015 4:57 pm

Re: Significance of impulse response VAR

Postby dakila » Sun Aug 07, 2016 2:24 am

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

Postby sofp » Sun Aug 07, 2016 2:43 am

Thank you so much!

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

Re: Significance of impulse response VAR

Postby soetirl13 » Fri May 18, 2018 8:34 am

Hi,

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

startz
Non-normality and collinearity are NOT problems!
Posts: 3376
Joined: Wed Sep 17, 2008 2:25 pm

Re: Significance of impulse response VAR

Postby startz » Fri May 18, 2018 9:11 am

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


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 4 guests