I am a student in finance and I am working on my thesis project. I am interested in doing a clustering to stock time series. I first read the paper 'Time-series clustering – A decade review' from Saeed Aghabozorgi, Ali Seyed Shirkhorshidi n, Teh Ying Wah on 'Information Systems'. But this paper provided too many methods and algos and many other choices such as representation method(Does stock time series needs to be transformed? I am thinking daily data of several years of the stocks in S&P 500, that's not a lot of data), Similarity measure approaches, prototypes,etc. And now I can't really decide what to do because I don't have experience on doing machine learning and right now I need to start somewhere. So anyone have indications on what to do next or which paper should I read next?

  • 1
    $\begingroup$ I don't think this is too broad as the question asks for literature to read, not a solution to the problem. $\endgroup$ – Peter Flom Sep 30 '18 at 10:37
  • $\begingroup$ @user598604 the paper you chose is a "review paper". You have to understand, a review paper collects all the past works and summarises it. I suggest choose a method that solves your research question and read related works on it. $\endgroup$ – mnm Oct 1 '18 at 4:24
  • 1
    $\begingroup$ You will have to try out a number of things. Don't expect an easy to use, single correct way! $\endgroup$ – Anony-Mousse Oct 1 '18 at 7:26

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.