Consider a medical test for a rare disease. Suppose the test is 99% accurate, but the disease affects only 1 out of 1,000 people. Many people assume that a positive test result means they almost certainly have the disease. A Bayesian analysis shows otherwise: because the base rate is so low, the probability of actually having the disease after a positive test is only about 9%. This large difference comes from properly weighing the evidence against the prior probability. | Consider a medical test for a rare disease. Suppose the test is 99% accurate, but the disease affects only 1 out of 1,000 people. Many people assume that a positive test result means they almost certainly have the disease. A Bayesian analysis shows otherwise: because the base rate is so low, the probability of actually having the disease after a positive test is only about 9%. This large difference comes from properly weighing the evidence against the prior probability. |