Any requests for what to include in a central limit theorem video? (Patreon)
Content
Hey everyone,
What began as the follow-on to the convolutions video has become a video about the central limit theorem. In particular, I'd like to show how a certain perspective on convolutions makes it easier to "see" why the function e^(-x^2) is the one that shows up so prominently in statistics, as opposed to some other function that bulges in the middle and tapers out in the ends.
The explanation is loosely connected to a delightful, and somewhat classic, argument for why the area under this curve is the square root of π, which in turn explains why the formula for a normal distribution has a π in it. Right now, the plan is to pair the main CLT explainer together with a second video talking more about why that pi is there, and how a kind of circular symmetry more core to this distribution than it might initially seem.
Given that pi day is coming up, it's only fitting to target publishing the two of these around that time.
In the meantime, I'm putting together a draft of the first one to share early here on Patreon, and as I do so I thought I'd ask whether any of you have any requests. For those of you who have seen the central limit theorem before, was there anything you found confusing upon first learning it? Do you have any favorite examples that helped to illustrate what it is saying? Do you have any lingering questions about the theorem, normal distributions, etc.?
For you're early viewing pleasure, here are a couple of simulations I rendered for the project. Let me know if you have thoughts for other such simulations in a similar vein.
Cheers,
Grant
Files
Previews only