Once upon a time there was a cute, little method called the Linear Regression. It had some interesting uses. It blew many people’s minds. It was loved and cherished by economists. But then some people loved it too much. They used it for everything. Everything. I’d like to give a little example (nonlinear relationship) and a possible fix for the problem.
I’m usually a Python user myself, but for quick and easy convenience, along with the fact that a lot more economists and social scientists use
You might see where this is heading. Let’s try running the simple linear regression
Well here’s a problem!!! In case you didn’t notice, the variable from which zzz is derived (xxx) does not show as "significant" (there’s a discussion for another day) in the linear regression. Oh, and yyy, the variable which is only related to zzz via xxx, shows as unbelievably significant. The only thing even close to right here is the standard error. To be fair, who can blame the computer?
Holy crap! yyy drops out like the poser it really is, while xxx makes a move to become zzz's new significant other (pun totally intended). And thus they lived happily ever after. And look at that smile!
I’ve considered working on a paper to similar effect as this blog post, obviously more involved, using previous studies, technical technicalities, and cool applications. Suggestions are appreciated!
Leave a Reply.
I'm the cofounder of a fintech startup