Bayesian Updating for true/false tests
Created on 2021-08-21T08:44:37-05:00
Statistical method of examining the prior probabilty of a condition, and determining the new probability given an imperfect test which has a known false positive and false negative rate.
- Sensitivity is the liklihood of a true positive
- Specificity is the liklihood of a true negative
PV+ = P(B|A) = \frac{P(A|B) P(B)}{P(A|B) P(B) + P(A|B') P(B')}
- B' is pronounced B Prime
- Prime in this case meaning the opposite of the probability
- The opposite of the prior is one minus the prior, ex. 1-x
- The opposite of the sensitivity is the specificity
- P(B) is the prior probability of having a condition
- P(A|B) is the sensitivity of the test