News for hackers. Stuff that matters.
Re:Datasaurus: Never trust summary statistics; visualize you... by nabla9 on Thursday August 27, 2020 @10:16AM (#24293914)
Eyeballing 2d plot is important statistical tool.

If it looks like a blob, correlation coefficient is usually meaningless. You can instantly see if the relationship is linear, quadratic, piecewise linear or quadratic, mixture, clustered or even dinosaur.

Re:Datasaurus: Never trust summary statistics; visualize you... by jdonaldson on Thursday August 27, 2020 @11:40AM (#24294753)
Except scatter plots do not handle occlusion. If you have a lot of data to plot, the blob shape might be outliers.
Re:Datasaurus: Never trust summary statistics; visualize you... by Petefine on Thursday August 27, 2020 @12:09PM (#24295083)
This might be a problem, but if you're careful to make use of transparency, axis scaling and random jitter (for integer or categorical values), the occlusion issue can be overcome.
Re:Datasaurus: Never trust summary statistics; visualize you... by jdonaldson on Thursday August 27, 2020 @01:14PM (#24295924)
* in some cases
Re:Datasaurus: Never trust summary statistics; visualize you... by nabla9 on Friday August 28, 2020 @07:41AM (#24303716)
Eyballing detects outliers as well. It can do density estimation.
Re:Datasaurus: Never trust summary statistics; visualize you... by SubiculumCode on Thursday August 27, 2020 @12:44PM (#24295528)
my scatters do. Transparency, dodge, etc.