Probabilities of Future Decisions

Today there exist efficient algorithms for finding the optimal strategy for decision problems formulated in an influence diagram. In this paper we construct a Bayesian network that enables the decision maker to examine the optimal strategy, or any other strategy, with respect to other criteria than maximal expected utility. We introduce the concept: ‘the probability of making a future decision’, and show how the constructed Bayesian network can be used to compute this probability. Other applications of the Bayesian network are shown, including methods for computing the risk profile of a strategy and the variance of the utility of a strategy.