0
$\begingroup$

I have two questions:

  1. Since VAR (vector autoregression) will not handle seasonality and trend. VECM comes into play which can handle season as well as trend. I had a doubt whether it will handle multiple seasons or not?

  2. Is there any other method to handle multiple seasonality in multivariate time series?

$\endgroup$
0
1
$\begingroup$

Since VAR(Vector Autoregressive) will not handle seasonality and trend. VECM comes into play which can handle season as well as trend.

This is incorrect. From the perspective of seasonality, VAR and VECM are fundamentally the same.

You can incorporate seasonal patterns via seasonal dummies. This works for multiple seasonalities, too. Just set up as many sets of dummies as there are seasonalities.

$\endgroup$
2
  • $\begingroup$ Thnx. What about the trend? Is there is any model/example for seasonal dummies in var? (preferably in python) If so please share $\endgroup$ – saravanan saminathan Oct 5 '18 at 13:54
  • 1
    $\begingroup$ You can also include a trend as an external regressor, no problem. Some examples for R are associated with the vars package (see vignette). I do not know about VARs or VECMs in Python. $\endgroup$ – Richard Hardy Oct 5 '18 at 15:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.