ABSTRACT
However, in most situations, the decision-maker expects his decision to have an effect on the state of the world, and she needs to solve the following problem, where conditional probabilities replace simple probabilities:
(10.2)
Typically, a decision-maker needs an answer to the question: ‘What happens if one decision is taken instead of another one?’. The estimation of conditional probabilities would therefore be a major qualitative advance relative to the estimation of simple probabilities. A policy-maker could, for instance, ask:
• What is the best policy to reduce crime? Tougher prison sentences? More police?
