Bayes rule on a medical test

Back to list2025-03-30
You

A test is 99% accurate and I tested positive. Am I 99% likely to be ill?

ChatGPT

No, and the gap is the whole point of the rule:

\[P(D \mid +) = \frac{P(+ \mid D)\,P(D)}{P(+ \mid D)P(D) + P(+ \mid \lnot D)P(\lnot D)}\]

With a prevalence \(P(D) = 0.001\) and both error rates at 1%, this gives roughly \(0.09\). Nine percent, not ninety-nine, because false positives are drawn from a pool a thousand times larger.