I want to try and predict previous values for random data that has properties of ARIMA process.

Problem is that I dont understand ARIMA model mathematically, so I dont know if that's even possible or allowed.

For example can I simply invert time axis and use ARIMA predictor? I would guess no.
Or can I use negative shift operator somehow?

  • $\begingroup$ If the process with inverted time scale still looks like ARIMA, I do not see a reason not to use ARIMA. $\endgroup$ – Richard Hardy Aug 18 '15 at 16:50

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.