I wanted some feedback on what is thought of as best practice in terms of showing correlation between two metrics in terms of minutes vs minutes when both metrics contain a large amount of data points ( hundreds of thousands or low millions trying to avoid a scatter plot with too many data points making it unreadable.


  • $\begingroup$ Density histograms are great for this. Please give an example data set with a small number of points, so that we can be sure what you mean, and propose concrete solutions. $\endgroup$ – Roman Jun 10 '19 at 21:16
  • $\begingroup$ Along the same lines, er, shapes that @Roman suggests is hexagon binning: mathematica.stackexchange.com/questions/28149/…. $\endgroup$ – JimB Jun 10 '19 at 22:01
  • $\begingroup$ @Roman Thanks for the response, adding example data now, is it possible to chart density histograms in Python/pandas? $\endgroup$ – Chris Jun 11 '19 at 4:41
  • $\begingroup$ @Chris you'll have to go elsewhere to ask about Python/pandas. This forum is about Mathematica. $\endgroup$ – Roman Jun 11 '19 at 4:51
  • $\begingroup$ Did you perhaps mean "matrices" where you wrote "metrics"? $\endgroup$ – Daniel Lichtblau Jun 11 '19 at 15:12

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.