Decision Analysis for Management Judgment, 6th Edition

9Revising judgments in light of new information

Introduction

Suppose that you are a marketing manager working for an electronics company which is considering the launch of a new type of pocket calculator. You estimate there is roughly an 80% probability that the sales of the calculator in its first year would exceed the break-even level. You then receive some new information from a market research survey. The results suggest that the sales are unlikely to reach the break-even level. How should you revise your probability estimate in light of this new information?

You realize that the market research results are unlikely to be perfectly reliable: the sampling method used, the design of the questionnaire, and even the way the questions were posed by the interviewers may all have contributed to a misleading result. Perhaps you feel so confident in your own knowledge of the market that you decide to ignore the market research and leave your probability unchanged. Alternatively, you may acknowledge that because the market research has had a good track record in predicting the success of products, you should make a large adjustment to your initial estimate.

In this chapter, we will look at the process of revising initial probability estimates in light of new information. The focus of our discussion will be Bayes’ theorem, which is named after an English clergyman, Thomas Bayes, whose ideas were published posthumously in 1763.¹ Bayes’ theorem will be used as a normative tool, ...