1
$\begingroup$

I estimated a VAR-model. I checked the time series for stationarity and after estimating the model the residuals and all is fine.

I know that it makes no sense to directly interpret the coefficients but I wondered if it is eligible to interpret the signs of the coefficients? E.g. The second lag is significant and seems to have a negative influence on y. ?

Thanks in advance.

enter image description here

Improve this question
$\endgroup$
2
  • 1
    $\begingroup$ That's why people typically look at impulse responses, which are easier to interpret. $\endgroup$
    – Christoph Hanck
    Commented Jan 8, 2021 at 15:13
  • $\begingroup$ To the extent that you cannot directly interpret the coefficient, the same likely holds for its sign as well. $\endgroup$
    – Richard Hardy
    Commented Jan 8, 2021 at 15:54

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.