Let's say you have some time series data that you think exhibits a drift, but you are unsure. Is it better to test first a regression with a drift and a trend to see if that is stationary?

Secondly, if it is stationary, then you know the drift is also stationary. If it is not stationary, then check only the regression with a drift.

Just looking for some guidance about how to approach and generalize the process for intermediate people like myself.

  • $\begingroup$ Have you tried differencing? $\endgroup$ – Math1000 Oct 6 '16 at 7:37
  • $\begingroup$ I was assuming some theoretical data. $\endgroup$ – Jack Armstrong Oct 6 '16 at 14:16

Your Answer

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

Browse other questions tagged or ask your own question.