Choosing Among Risk Profiles

Abstract

To solve risk management problems using the expected utility (EU) framework, one must first represent different decision options by their risk profiles using QRA modeling methods such as those in chapters 2–4. Then, it is necessary to choose the option corresponding to the most-preferred risk profile. This chapter covers methods for choosing among risk profiles — or, equivalently, cumulative probability distribution functions (CDFs) — when the consequences of interest can be summarized and compared using a single numerical attribute. This is not necessarily a severe restriction, as the attribute may be a synthetic one such as “net benefit” or “value” that summarizes preferences for consequences. The results in this chapter can be used to construct explicit utility functions for single-attribute consequences so that the expected utilities of prospects can be calculated. They also allow some pairs of prospects to be compared even without defining an explicit utility function, using various dominance criteria.