Breakthroughs in Decision Science and Risk Analysis

Chapter 7Making Decisions without Trustworthy Risk Models

Louis Anthony (Tony) Cox, Jr.

Cox Associates, NextHealth Technologies, University of Colorado-Denver, Denver, CO, USA

Challenge: How to make Good Decisions without agreed-to, Trustworthy Risk Models?

How can risk analysts help to improve policy and decision-making when the approximately correct relation between alternative acts and their probable consequences is unknown? This practical challenge of risk management with model uncertainty, which was discussed in Chapter 4 from the standpoint of learning how to act while recognizing that future information may cause current models to be revised or replaced, arises in even more extreme form when learning by trial and error is prohibitively costly or potentially fatal. Yet, decision-making without the support of models that can be trusted to deliver at least approximately correct results is a frequent feature of modern life, in problems ranging from preparing for climate change to managing emerging diseases to operating complex and hazardous facilities safely.

This chapter reviews constructive methods for improving predictions and decisions when the correct description of the causal relation between decisions and outcome probabilities is unknown or is highly uncertain. These methods are not yet as familiar to many risk analysts as older statistical and model-based methods, such as the paradigm of identifying a single “best-fitting” model and performing sensitivity analyses for its ...