Handbook of Decision Analysis

CHAPTER TEN

Quantify Uncertainty

ERIC R. JOHNSON and STEVEN N. TANI

Do not expect to arrive at certainty in every subject which you pursue. There are a hundred things wherein we mortals … must be content with probability, where our best light and reasoning will reach no farther.

—Isaac Watts

10.1 Introduction

10.2 Structure the Problem in an Influence Diagram

10.3 Elicit and Document Assessments

10.3.1 Heuristics and Biases

10.3.2 Reference Events

10.3.3 Assessment Protocol

10.3.4 Assessing a Continuous Distribution

10.3.5 Conditioning Cases

10.3.6 The Reluctant Expert

10.4 Illustrative Examples

10.4.1 Geneptin

10.5 Summary

Key Terms

References

10.1 Introduction

Our aim is to help decision makers identify choices that can add substantial value when facing complex, uncertain, novel, high-stakes decision situations. Our decision analysis approach is to develop a composite perspective on how value is created based on the best available expertise, and to employ it in a value dialogue with experts, stakeholders, and decision makers aimed at providing insight to decision makers to aid their decision making.

We develop the composite perspective probabilistically, but not in the spirit of resignation indicated by the Isaac Watts quote above. Like Watts, we acknowledge that perfect knowledge is almost never available; but we rejoice in the availability of expertise, and set out to marshal it and bring it to bear on the decision problem at hand.

Previous chapters discuss framing and structuring ...