0

How do I decide on the right number of bins to discretize my continuous data? Are there are tests/techniques to do the same? Could someone give me some idea into existing approaches?

AnonymousMe
  • 195
  • 4
  • 1
    Why do you want to use discretization at all? – Tim Dec 05 '21 at 07:57
  • Because, I have to compute a metric called conditional entropy rate which can only be done on discrete data. – AnonymousMe Dec 05 '21 at 07:58
  • 1
    Why do you have to compute a metric that relies on discretization? Discretization is almost always a bad idea: https://stats.stackexchange.com/q/68834/1352 If your KPI requires it, it would be a good idea to find out *why* that KPI should be used. – Stephan Kolassa Dec 05 '21 at 08:47
  • What is your purpose? Can you say what kind of data you have? – BruceET Dec 05 '21 at 17:16

0 Answers0