I have many items with a start date, which produce continuous outputs over time for a few months after the start date. These outputs often follow a decay function but not always. So it is a bit like a survival analysis problem. But rather than predicting a probability over time I need to predict a continuous value over time (resolution month). Which (machine learning) modelling could I use for this please? Thanks!

  • $\begingroup$ would you please post a link to an example that clearly has a decay-like function? $\endgroup$ – James Phillips Dec 7 '18 at 20:14
  • $\begingroup$ @JamesPhillips - the values do not always follow a decay function. But if you ask for what a decay function, I googled an example for you: mathworld.wolfram.com/ExponentialDecay.html $\endgroup$ – cs0815 Dec 8 '18 at 6:42
  • $\begingroup$ Your generously googled data example cannot help answer the question regarding your specific data. $\endgroup$ – James Phillips Dec 8 '18 at 13:39
  • $\begingroup$ @JamesPhillips you must understand that one cannot always upload data here. imho the problem is quite well described. $\endgroup$ – cs0815 Dec 8 '18 at 13:41

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.