Tutorial

Conditional Probabilities and Bayes Theorem

TL;DR Bayes Theorem lets you flip a conditional probability — calculate (P(A \mid B)) when you only know (P(B \mid A)), (P(A)), and (P(B)). The classic medical-test example: even with a 95% accurate TB test, if the disease is rare, a positive result means less than a 2% chance of actually being infected. I've been getting a lot of questions from friends lately about what Bayes Theorem means. The confusion is understandable because it appears in a few models that seem to be completely unrelated to each other.