A Contingency Framework for Risk Analysis

The paper presents, justifies, and illustrates a new approach to risk analysis. It is assumed that the conditional mean and variance of a random variable, the system effectiveness measure, can be determined for each contingency in the risk environment of the system under analysis. A contingency is a combination of discrete factor states with a given probability of occurrence. Exact formulas are presented for the overall mean and variance of the effectiveness measure. As well, formulas are presented for variance decomposition which allow the overall system variability to be split into contributing components. The approach is illustrated with a numerical profit budgeting example.