Problem_Session_7

Problem_Session_7 - MS&E 252 Decision Analysis I...

Info iconThis preview shows pages 1–12. Sign up to view the full content.

View Full Document Right Arrow Icon
Decision Analysis I Problem Session 7
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
What concepts do we expect you to master? Decision Diagrams Probability Encoding 20 Questions Quantitative Problem A quick overview of MS&E 352 – DA2
Background image of page 2
Why are decision diagrams important? Facilitates communication. Grows linearly, unlike trees that grow exponentially. No degrees and no probabilities yet! Therefore no clarity test required Enables high level abstraction
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Decision diagrams contain four kinds of nodes. Decision Node Uncertainty Node Deterministic Node Value Node F(x,y) x y
Background image of page 4
For example, in Kim’s party problem… Uncertainty Sunny or rainy weather? Value Value to Kim ($) Decision Where to have the party? Type of Node Corresponds to
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Four kinds of arrows indicate conditioning. Informational Arrow Relevance Arrow Functional Arrow Influence Arrow A B C D 2 D 1 E D F D C “B is conditioned on A, and may be relevant to A.” “C and D 1 are known when D 2 is made.” “F is a deterministic function of D and E.” “My choice in decision D may affect the distribution of uncertainty C.”
Background image of page 6
Let’s draw the decision diagram for Kim’s party problem. Kim decides without knowing the weather Kim’s value = fcn (W,L) $ ? Kim’s decision does not modify the weather Weather Location Watch out for “rookie mistakes”!
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Let’s try it on another example! DA Corp. is thinking of investing in research on a new drug, “Clear Voyance”. “Clear Voyance” would instantly and painlessly cure any eyesight problem. They estimate that if they decide to invest in research, technical success will only be 20% likely. However, once they have technical success, they have a 90% chance of marketing success. DA Corp. also needs to decide whether they should proceed to preliminary testing that will give an indication on the chances of technical success. How would you set up a decision diagram for this example?
Background image of page 8
Here is a possible decision diagram for DA Corp.’s decision $ Marketing Success Proceed to Tests Report Technical Success Test Indication Invest in Research Exercise – Name all the kinds of arrows on this diagram
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
What concepts do we expect you to master? Decision Diagrams Probability Encoding 20 Questions Quantitative Problem A quick overview of MS&E 352 – DA2
Background image of page 10
Why do we need probability encoding? Probability encoding is a process
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 12
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/16/2010 for the course MS&E 252 taught by Professor Howard during the Fall '08 term at Stanford.

Page1 / 33

Problem_Session_7 - MS&E 252 Decision Analysis I...

This preview shows document pages 1 - 12. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online