Sensitivity of Decisions to Probability Estimation Errors

Abstract

In making decisions on the basis of maximum expected utility, the decision maker is required to analyze the sensitivity of his decision to variability in probability estimation. This paper explores the general question of sensitivity to subjective probability estimates and derives an analytical representation of sensitivity. Some of the problems of applying sensitivity analyses are explored and a method suggested for resolving conflict of choice under conditions of high sensitivity. A sample problem is provided to illustrate application of the analytical techniques.