Occasionally, you will encounter situations in which you assess conditional probability information one way (e.g., what is the probability that A is high given that B is high) but encounter the events chronologically in the reverse order (i.e., given B is high what is the probability A is high). For example, the results of a screening test used to detect a disease depend on whether or not the tested person is infected, but you know the results of the test before you know if the person is truly infected. In DPL, you would draw a green arrow (indicating probability conditioning) from a node called Patient Infected to a node called Test Results in the influence diagram. In the decision tree, Test Results would appear before (to the left of) Patient Infected.
The probabilities in the tree must be calculated using a technique called Bayesian revision (using Bayes' Rule). DPL handles this automatically and puts the calculated probabilities on the rolled-back tree.
Versions: DPL Professional, DPL Enterprise, DPL Portfolio