I have financial time series and PCA scores, that I'm trying to cluster. As PCA scores don't have orientation, I would like to know what clustering method would be suitable for clustering these kind of series? I feel that -1 correlation series with my PCA scores is as important as +1 correlation and should be clustered together.

  • $\begingroup$ I would use k-means. $\endgroup$ – james42 Nov 22 '16 at 9:45
  • $\begingroup$ Thank you for you comment, is there preferred similarity matrix method? I have used correlations, but there is alot to choose from (wavelet, DTW)? $\endgroup$ – Viitama Nov 22 '16 at 9:52
  • 1
    $\begingroup$ I'm voting to close this question as off-topic because this is a purely statistical question and as such would better fit at Cross Validated. $\endgroup$ – Richard Hardy Nov 22 '16 at 19:06
  • $\begingroup$ Sure, thought that it would fit under Quantitative Finance, but cross validated might be better then. Thanks for effort. $\endgroup$ – Viitama Nov 23 '16 at 8:00

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.