Decision Modeling and Risk Analysis in Introductory Statistics
David Doane
Department of Decision and Information Sciences
Oakland University
Rochester, Michigan 48309-4493
doane@oakland.edu
KEY WORDS: simulation, decision modeling, distributions, @RISK
Many students doubt that statistical distributions are of practical value in business. The simulation capabilities of @RISK and similar software make it possible for students to tackle challenging yet understandable projects that illustrate how distributions can be used to improve business decisions and answer "what-if" questions of the type often posed by managers. Course materials that have been developed over two years of classroom trials will be shared, including (1) an in-class demonstration to allow the instructor to illustrate the capabilities of @RISK; (2) two different in-class @RISK spreadsheet projects designed for teams of 2 students; and (3) handouts and instructions that pose specific questions to be answered by the teams. These materials can easily be revised to ensure fresh questions. Concepts illustrated include expected value, k-tiles (e.g., quartiles), empirical distributions, distribution parameters, and the law of large numbers. For the benefit of those who don't have @Risk, a spreadsheet will be shown in which many risk-modeling concepts are accomplished using only Excel. Live demonstrations will be given.