I have three data sets from same source. I know they are interdependent and stochastic also. The datasets are of very high frequency (see the graph, X axis is time and Y axis is value). I want to represent these three signals in a single mathematical function. Is it possible? enter image description here

In the above graph, please see there is a pattern or seasonality. If we change one then other also changes. I have tried to google for it but did not find any model that can completely represent this type of data in a single expression.

  • $\begingroup$ Welcome to DSP.SE! Mathematical models of signals are very easy to formulate. However, you've said nothing that would allow us to help you with this. Please let us know: Are the signals sine waves? Are they stochastic (random)? Is there any structure at all? What do you mean by "interdependent"? What do you mean by "very high frequency"? $\endgroup$ – Peter K. Sep 22 '15 at 15:59
  • $\begingroup$ I have edited @Peter-K Hope now it is clearer. $\endgroup$ – Kal Sep 22 '15 at 16:59
  • $\begingroup$ Thanks! I've reopened. Let's see what people come up with. $\endgroup$ – Peter K. Sep 22 '15 at 17:00
  • $\begingroup$ What exactly do you want the function to do? $\endgroup$ – Jim Clay Sep 22 '15 at 17:29
  • $\begingroup$ @JimClay : I want a mathematical expression(s) that represents these three stochastic signals, in other words if I give the input (say we have three signals a,b,c so if we input the value of a,c) then it will produce the result of b at that time quantum. $\endgroup$ – Kal Sep 22 '15 at 17:48

Try the Eureqa software. It tries to find the simplest mathematical relation between multiple data. If the relation truly has a simple symbolic form, I think it should find it. Otherwise it will produce some strangely formulated approximation.

  • $\begingroup$ Thanks!! At a first glance it looks like an appealing tool ... $\endgroup$ – Kal Sep 23 '15 at 9:40
  • $\begingroup$ Eureqa gives me some strange equation when I tried it to fit other set of data it does not fit well (data from same source but on different time).. $\endgroup$ – Kal Sep 25 '15 at 10:40

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.