EViews.com

I have a VAR model with logarithmic data with just one endogenous variable, I had 2 dummies, but their coefficients apeared to be statistically insignificant. Anyways, I thought at least I could do the impulse response analysis. How could you interpret this graph? Second graph is with original data with special events, which could have done (as a hypothesis) some impact .

I would appreciate your help! I feel like I need some thoughts from the side.

Thanks in advance.
Eviews7

the dotted / stippled lines show the standard error by +/- 2
let's say your data is in daily observations, you could in this way interpret the x-axis as number of days into the future.
one could say the effect of one sd shock puts the response variable out of it's equilibrium and does not seem to return in the specified time period.

not sure if this is helpful, but there is a guy Sayed Hossein who does wonderful Eviews tutorials on youtube, i recommend you look.
for everyone else, if something is wrong, please correct