Probabilistic Causality and Its Applications to Risk Analysis

Abstract

Causal notions play an integral role in almost every facet of the risk analysis profession. In particular, risk analysts are typically interested in two types of causal relations. In the first type, the objective is to determine the tendency of one event to cause another. In the second type, the objective is to determine which cause, among a set of potential causes, is the most dominant contributor to an observed effect. This paper examines various theories of probabilistic causation, with particular emphasis on Patrick Suppes’ theory. The potential relevance of probabilistic causality to various risk analysis applications is then considered.