I have a system producing an output periodically, I would like to build a model to predict the next entry.

There is no sequence relation between the output values, order doesn't matter. The only thing is that output is most likely to be a previous value or something close to it.

as an example: 5, 3, 6, 5.2, 5.4, 4.9, 3, 8

this model would predict a value around 5 given that most of the values are in that range.

Apologies for the vague explanation, I lack the correct terminology.

  • $\begingroup$ Does the sequence of outputs change at all over time, or is it always within a similar range? $\endgroup$ – rw2 Jan 10 at 11:45
  • $\begingroup$ It would take all of them, $\endgroup$ – Mhd.Tahawi Jan 10 at 11:46

It seems like you are trying to determine a probability distribution from a set of examples. This would correspond to distribution fitting, if a family of possible distributions is known. In most statistics software this would correspond to intercept-only models with various response distributions.

The problem could also be understood as quantile regression with only the intercept.

Or you could just construct the empirical distribution function and use this to form your predictions.

But in general I think you would get better results by trying to find some constraints on the process that generates your values and use distribution fitting or other not extremely flexible methods.

Hope this gets you started.

(edited, originally I misunderstood the question).


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.