This repo consists of a series of notebooks exploring the statistical underpinnings of risk-limiting audits. I start with fairly rudimentary structures and walk through developing enough statistical machinery to tackle the main types of RLAs.
The notbooks are ordered:
This is all still a work in progress, so things will be rough for a while. I'm not a statistician™, so a lot of this is me fumbling around in the dark to try to understand how it all works. In the event that that's useful to someone else, I hope this helps!