Improve hexbin() documentation #20397
Conversation
timhoffm
commented
Jun 9, 2021
timhoffm
commented
- See-also-link between hexbin() and hist2d()
- More compact example code notation
- Better example Introduction
examples/statistics/hexbin_demo.py
Outdated
| Hexagonal bins have less sampling bias from edge effects than rectangular | ||
| bins and can thus be a more correct representation of the data. |
There was a problem hiding this comment.
| Hexagonal bins have less sampling bias from edge effects than rectangular | |
| bins and can thus be a more correct representation of the data. | |
| For some data distributions, hexagonal bins may have less sampling bias from edge effects than rectangular bins. |
... though I'd just not say this at all, unless you have a citation to say that hexagonal bins are "more correct"
There was a problem hiding this comment.
Hexagons are the maximum number of sides a polygon can havefor a regular tesselation of the plane, so in terms of packing a hexagon is 13%more efficient for covering the plane than squares. This property translates intobetter sampling efficiency at least for elliptical shapes. Lastly hexagons are vi-sually less biased for displaying densities than other regular tesselations. Forinstance with squares our eyes are drawn to the horizontal and vertical lines ofthe grid.
https://cran.r-project.org/web/packages/hexbin/vignettes/hexagon_binning.pdf
There was a problem hiding this comment.
I agree for the example shown there, where the size of the bin is proportional to the density in the bin. Less convinced that it is true for a normal hex bin plot.
timhoffm
force-pushed
the
doc-hexbin
branch
2 times, most recently
from
ba4a3ec
to
9361a71
Compare
- See-also-link between hexbin() and hist2d() - More compact example code notation - Better example Introduction Co-authored-by: Jody Klymak <jklymak@gmail.com>
