Animating causal inference methods

Some time back the animations below went sort of viral in the statistical programming community. In them, economics professor Nick Huntington-Klein demonstrates step-by-step how statistical tests estimate effect sizes.

You will find several other animations in Nick’s original blog, and the associatedtwitter thread.

Moreover, if you are interested in the R code to generate these animations, have a look at this github repository for the causalgraphs.

Controlling for a variable

Via http://nickchk.com/causalgraphs.html

Matching on a Variable

Differences in differences

Link to the Twitter thread:

I've been getting used to gganimate and thought it would be useful to put together some illustrations of what various causal inference methods *actually do to data* and how they work. Here, for example, is what it means to control for a (binary) variable pic.twitter.com/lmEvJSPQgY
— Nick HK (@nickchk) November 26, 2018

One thought on “Animating causal inference methods”

Super cool. Not sure I understand the methods but it appears to be some kind of regression, at least in part. I’ve been working on graphics in r but this is way more advanced than what I’m doing. Thanks for sharing!

LikeLike