I have a tuple that looks like this when I plot the graph: the tuple itself But the minima are not quite clear. So I tried to use logarithmic scale and visualise the log of the tuple instead of the tuple itself. log of the tuple But now, the maxima are not clear. Eventually I tried -1/log(x). This way both maxima and minima look clear. -1/log But I wonder if this is the neatest way to scale that type of data. Would you recommend any other scaling options?

  • $\begingroup$ Is this a time series? Please tell us what this data represents! $\endgroup$ Sep 11 at 11:19
  • 1
    $\begingroup$ no, this is a "consonance" (harmoniousness of two musical tones) plot within a span of one octave (the range between two tones with frequency ratio of 2) the x axis is the 12 tones within the octave, the y axis is the harmoniousness of those 12 tones with the first note. $\endgroup$
    – artmyb
    Sep 11 at 11:33
  • $\begingroup$ Please add that explanation as an edit to the post! It might help somebody to give an informed answer.- $\endgroup$ Sep 11 at 11:40
  • 1
    $\begingroup$ Try cube root scale. Logarithmic scale is clearly too strong in that extreme minima pop out as outliers. I wish I understood the science here to be less ad hoc. $\endgroup$
    – Nick Cox
    Sep 11 at 12:31

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.