I'm getting conflicting information on whether stationarity is a requirement when using neural networks for time series forecasting:

(These aren't the only two sources where I've seen this conflicting information).

Can someone help me figure which is correct? Or if the answer is ambiguous (i.e. still an open question), can someone point me to a good resource (review paper, book chapter, etc...) on the topic?

  • $\begingroup$ Can you specify which parts of the lectures talk about it? $\endgroup$ – vpekar May 9 '19 at 21:34
  • $\begingroup$ @vpekar in the first lecture, it is somewhere in the last 10 minutes (When he starts talking about LSTM). In the second video, he starts talking about it after the 9th minute. $\endgroup$ – Akaike's Children May 9 '19 at 21:39

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.