In this chapter you will learn about an approach to decision making called Decision Analysis (note the capital letters) that is based on estimates of the probabilities of each state of nature. It is also based on measures of the overall desirability of each outcome, called its utility, so ultimately it is a one-criterion approach. The structure of the decision making problem is represented as a decision tree. You will examine the sensitivity of decisions to different assumptions about the probabilities involved. Finally, there is a brief introduction to decision conferencing, in which Decision Analysis is used by groups.
By the end of this chapter, you will be better able to:
• represent the structure of a decision making problem as a decision tree • calculate the expected utility of different decision options • carry out a sensitivity analysis based on a decision tree • describe the method of decision conferencing for group decision making
Chance node A point at which a decision tree branches into a mutually exclusive set of states of
nature. A chance node is usually represented by a circle.
Decision Analysis An approach to decision making which involves representing the problem
in decision-tree form, branching at decision nodes and chance nodes. Probabilities are needed
for each branch out of a chance node and utilities for each final branch in the tree.
Decision node The point in a decision tree where a decision must be made between competing
and mutually exclusive policy or treatment options.
Decision tree A type of model of a decision making problem with branches representing the
possible decision options and states of nature.
Expected utility The benefit or satisfaction that an individual anticipates getting from
consuming a particular good or service.
Lottery A hypothetical gamble used in Decision Analysis to estimate the utility of an outcome.
Decision outcome A combination of decision options and states of nature. Each 'terminal'
branch of the decision tree represents an outcome.