8 Replies to “ggplot2 awesomeness at play!”

  1. +James Holland because people often use them to show their data, and in that context what they do is hide the data – e.g. how much there is and exactly where. Particularly problematic at the edges of a data set. For example here it could be significant that there are no red points < 21 mpg; maybe the expt went wrong; the density curve obscures that.

    Density estimation is an inference or fitting procedure, so fine to use it in that context.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.