QR 26: CHOICE AND CHANCE

Assignment # 2: Constructing Decision Trees and Diagrams

1. Study Chapters 3 and 4 in Making Hard Decisions (MHD). While it is good to know how to make and read "influence diagrams," you do not have to pay them much attention beyond that since we will concentrate on using the more traditional decision trees. Compare MHD with the excerpts in our Sourcebook from Chapters 1 and 2 of Clinical Decision Analysis by Weinstein and Fineberg.
2. Begin using the PrecisionTree Add-On for Excel to draw and solve decision trees. The software and tutorials are accessible from Windows machines in computer laboratories. Using Internet Explorer (not Netscape), just start from the link at the bottom of the QR26 web page. Videos and demos there can help you learn PrecisionTree step by step. (These tutorials are under development by Professor Terrence Reilly, who has kindly made them available for use in and by this class only, with the hope of obtaining constructive feedback from all of us.) The tutorial web pages can also be read from any browser on campus, but without the video and animation. TreePlan, a basic alternative to PrecisionTree that runs on Windows or Macintosh machines, is available through course servers and is linked to the handouts section of our web page.
3. Problem Set from MHD to hand in:

• At least one of #3.9 or 3.10.
• #3.20
• At least one of #3.22, 3.23, or the "Prescribed Fire" Case (p. 106).
• #4.15

1. Prepare notes that you can discuss and hand in concerning:

• "Job Offers Case" (pp. 169-170). Try to analyze this also using the "Method of Equal Swaps" you read about in Smart Choices. Extra Credit: Look up and critique how magazines actually do rate the desirability of different cities as places to live.

1. Start thinking about how you could explain what these probabilities we are using actually mean. (You do not have to hand in anything about this yet.) Where do they come from? What do we do with them? What properties must the numbers satisfy and why? How can you assign probabilities to an infinite number of possible events?