Persistence and probabilistic projection

Predicting the future is an essential component of decision-making. In most situations, however, there is not enough information to make accurate predictions. A theory of causal reasoning for predictive inference under uncertainty is developed. A common type of prediction that involves reasoning about persistence is emphasized: whether or not a proposition once made true remains true at some later time. A decision procedure with a polynomial-time algorithm for determining the probability of the possible consequences of a set of events and initial conditions is provided. The integration of simple probability theory with temporal projection circumvents problems in dealing with persistence by nonmonotonic temporal reasoning schemes. These ideas have been implemented in a prototype system that refines a database of causal rules in the course of applying those rules to construct and carry out plans in a manufacturing domain. >