Given that I select features manually, what methods are available for pattern discovery with the purpose of time series prediction (footnote)?

I only stumbled upon hierarchical clustring ("bottom-up") and proprietary sofftware. This post says there are a lot such methods.

Footnote: prediction in wider sense is meant. The outcome to predict is chosen manually, too, from the trivial "color of the next candle stick" via "wiskers will be longer than the body" to "next day will set a higher high", just anything, you name it


Here's a few:

  • K-means
  • PCA
  • HMM (learned with expectation-maximization or Viterbi algorithm)
  • Autoencoders
  • Outlier detection, e.g. Chauvenet's criterion
  • $\begingroup$ picking KNN as example because of its simplicity (distance measures aside..), how's prediction happening AND how are patterns discovered? $\endgroup$ – Xpector Jan 16 '18 at 17:35
  • $\begingroup$ Strictly speaking, KNN is supervised because you give it training examples. You can set it up so that it's not trained against the actual labels you are trying to predict. $\endgroup$ – madilyn Jan 16 '18 at 17:45

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.