0
$\begingroup$

I need to classify data that changes over time. A good example is stock prices. My problem is that stock prices examples are trying to predict the next price and I need to classify the data. I looked at text classifiers but they are based on human text. Another example is credit score classification in which the data is income and expenses of people. The target is to classify the data to groups. It can be reinforces learning as each sample belongs to a woman/man/couple/young/aged, etc. Or it can be that the classifier finds new classes.

So can anyone help with a code example/references on how to classify data that changes over time?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

Testing for mean change and/or variance change is a suggested approach to classify time series data. Understand order of time series discusses this. A third way to aid the classification of a time series is to identify points in time where the parameters of the model change.

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thanks. The problem is not how to analyze the data. It is to find the an algorithm that will do classification rather then prediction. Maybe it is not the right place to ask... $\endgroup$
    – nmnir
    Commented Sep 30, 2018 at 18:01
  • $\begingroup$ All that I said is useful to classify ...no prediction is necessary or in your case wanted. $\endgroup$
    – IrishStat
    Commented Sep 30, 2018 at 20:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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