Chapter 12, Value of Information by hd3hIc57

VIEWS: 3 PAGES: 21

									               Chapter 12
• Value of Information




                            1
Chapter 12, Value of information
•   Learning Objectives:
•   Probability and Perfect Information
•   The Expected Value of Information
•   Expected value of Imperfect Information
•   Value of information in Complex Problems
•   Value of information and Experts

                                               2
Chapter 12,Value of Information
• Decision Maker often gather information to
  reduce uncertainty
• Information gathering includes:
  – Consulting experts, conducting surveys
  – Performing mathematical or statistical analysis
  – Doing research, or simply reading books,
    journals, and newspapers.

                                                      3
        Value of Information
• Value of Information: Some Basic Ideas
• Probability and perfect information
• Use conditional probabilities and Bayes’
  theorem to evaluate information in any
  decision setting.



                                             4
        Value of Information
• The Expected Value of Information
• By considering the expected value, we can
  decide whether:
  – An expert is worth consulting
  – Whether a test is worth performing
  – Or which of several information sources would
    be the best to consult.

                                                    5
       The Expected Value of
            Information
• The worst possible case:
• Regardless of the information we hear, we
  still would make the same choice that we
  would have made in the first place.
• In this case, the information has zero
  expected value.


                                              6
       The Expected Value of
            Information
• Make a difference choice, then the expected
  value of the information must be positive
• The expected value of information can be
  zero or positive, but never negative.
• Different people in different situation may
  place different values on the same
  information

                                            7
    Expected Value of Perfect
          Information
• The optimal choice in any decision making
  situation is the one with the highest
  Expected Monetary Value (EMV)
• How much would he be willing to pay for
  information that would help you to make
  the right decision?


                                              8
     Expected Value of Perfect
           Information
• To find the value of these information, find
  the EMV for each situation and then
  subtract them.
• We can interpret this quantity as the
  maximum amount that the investor should
  be willing to pay for perfect information.


                                                 9
   Expected Value of Imperfect
          Information
• We rarely have access to perfect
  information.
• In fact, our information sources usually are
  subject to considerable error.
• Thus, we must extend our analysis to deal
  with imperfect information.
• We still consider the expected value of the
  information before obtaining it, and we will
  call it the (EVII).                          10
Value of Information in Complex
            Problems
• In most of previous example there was only
  one uncertain event
• Most real-world problems involves
  considerably more complex uncertainty
  models.
• In complex situation we must consider two
  specific situation

                                           11
Value of Information in Complex
            Problems
• First how to handle continuous probability
  distribution
• Second, what happen when there are many
  uncertain events and information is
  available about some or all of them
• Evaluate decision option with and without
  the information, and find the difference in
  the EMV

                                                12
        Value of Information
• Sensitivity Analysis, and Structuring
• The first step is using a tornado diagram,
  those variables to which the decision was
  sensitive.
• The second step, after constructing a
  probabilistic model, may be to perform
  sensitivity analysis on the probabilities.

                                               13
         Value of Information
• A third step in the structuring of a
  probabilistic model would be to calculate
  the EVPI for each uncertain event.
• If EVPI is very low for an event, then there
  is little sense in spending a lot of effort in
  reducing the uncertainty by collecting
  information.

                                                   14
         Value of Information
• But if EVPI for an event is relatively high, it
  may need to collecting of information
• Such information can have a relatively large
  payoffs by reducing uncertainty
• This information can also improving the
  decision maker’s EMV.


                                               15
        Value of Information
• Value of Information and Nonmonetary
  Objectives
• In most cases the only objective that matters
  is making money
• However in many decision situations there
  are multiple objectives.
• For example, consider the FAA bomb-
  detection case again.
                                              16
        Value of Information
• FAA was interested in maximizing the
  detection effectiveness and passenger
  acceptance of the system
• while at the same time minimizing the cost
  and time to implementation.
• Minimizing cost happens to be one of the
  objectives.

                                               17
        Value of Information
• The answer would be to find the additional
  cost
• Additional cost make the net expected value
  of getting the information equal to the
  expected value without the information
• Trade-off always establish to value the
  information

                                            18
        Value of Information
• Suppose one objective is to minimize the
  decision maker’s time
• Different choices and different outcomes
  require different amounts of time from the
  decision maker.
• Information can be valued in terms of time;


                                            19
        Value of Information
• Value of Information and Expert
• Expert information typically is somewhat
  interrelated and redundant.
• The real challenge in expert use is to
  recruiting experts who look at the same
  problem from very different perspectives.
• Use of expert from different field

                                              20
        Value of Information
• Summary
• Make better decisions by considering the
  expected value of information
• Both influence diagrams and decision trees
  can be used for calculating expected values
• How to solve value-of-information
  problems in more complex situations

                                                21

								
To top