I have multiple repeats of a time series that I would like to use to train a model to predict future repeats. The time series contains feature data (easy to measure) and target data (hard to measure). I would like to use the feature data to predict the target data. Is Kalman filtering a suitable approach for this? I would be interested in doing this using python.

  • 1
    $\begingroup$ Kalman filter can track non-stationary time series and estimate certain hard to observe quantities, but given the little information in the question it is hard to tell $\endgroup$
    – Nikos M.
    Nov 24 at 17:59
  • $\begingroup$ @NikosM. I can add I am currently approaching the problem in 2 methods. 1) An LSTM using the feature data to predict the target data. 2) A regression problem. $\endgroup$ Nov 24 at 18:45
  • $\begingroup$ Ok but this does not say anything about the problem itself, except that it is a time series $\endgroup$
    – Nikos M.
    Nov 24 at 18:53

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.