Explain Bayes' Theorem and walk through a practical example.
StatisticsLevel: IntermediateStatistics and probability
Explain Bayes' Theorem and walk through a practical example.
Hints
- Name the events and the conditioning information.
- Distinguish P(A|B) from P(B|A); check the event's base rate.
Answer
Bayes' theorem: P(A|B) = P(B|A) · P(A) / P(B). It lets you update a prior belief P(A) given new evidence B. Classic example: a disease affects 1% of a population; a test is 99% sensitive and 95% specific. If someone tests positive, P(disease|positive) = (0.99 × 0.01) / (0.99×0.01 + 0.05×0.99) ≈ 16.7% — much lower than intuition suggests, because the false positive rate dominates when the base rate (prior) is low.
Follow-up: Why is this base-rate fallacy dangerous in real-world screening/fraud systems?
Source and licence: Ishan Dutta — Awesome Data Science Interview Q&A · MIT