Visualizing Multivariate Data with Chernoff Fish

Humans are innately bad at understanding high dimensional data. After all, our world only includes three dimensions. If you don’t believe me—think about what a four dimensional sphere looks like. (Then check out this gif.)

This dimensional deficit impairs our ability to make informed decision about important everyday things, like finance. How you compare two mutual funds—when you look past their rate of return—becomes like comparing two unknown foods by their nutrition facts and ingredients.

What if you used an already recognizable object seen in nature to describe fund characteristics?

In Visualizing Financial Data, Julie Rodriguez and Piotr Kaczmarek explore just that with Chernoff Fish, a variation of Chernoff faces. Each part of the fish is used to encode different fund traits: pupil size maps to performance, the side and tail fins describe sector exposure, etc.

Chernoff Fish

I thought the concept was really cool, but it wasn’t interactive (yet). I couldn’t build my own fish! So I decided to make my own Chernoff Fish generator AND learn how to use React and D3 together.


You can check out the final result right here! (I recommend going crazy on the Randomize button.)


All my code is on GitHub.

Once-A-Month Thoughts Join 30+ subscribers