SlideShare a Scribd company logo
1 of 11
Download to read offline
I .
                                                                                                                                                                                                                                                                    I .
                                                                                                                                                                                                                                                                    !
                                                                                                                                                                                                                                                                    I
                                                                                                                                                                                                                                                                    j
                                                                                                                                                                                                                                                                    !



                                                                                                                                                                                                                                                                    .
                                                                                                                                                                                                                                                                    ! .
                                                                                                                                                                                                                                                                    f :
                                                                                            DECISION THEORY

DEFINITION 1.1 :

DECISION THEORY (DT) is a set of concepts, principles, tools and
techniques that aid the decision maker in dealing with compl~x decision
problems under uncertainty.

                                                                                                                                                                                                                     .-
COMPONENTS OF A DT PROBLEM:

1. THE DECISION MAKER

2. ALTERNATIVE COURSES OF ACTION
     This is the controllable aspect of the problem.

3. STATES OF NATURE OR EVENTS
     These are the scenarios or states of the environmen.tnot under the
                        control of the decision maker. The events defmed should be
                       mutually exclusive and collectively exhaustive.

4. CONSEQUENCES
     The consequences that must be assessed by the decision maker are
     measures of the net bertefit, payoff, cost or revenue received by the
                        decision maker. There is a consequence (or vector of
                        consequences) associated with each action-event pair. The
                        consequences resutilmarized a decisionmatrix.
                                   a              in                                                                                                                                                                                                           .




1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111II


DECISION THEORY                                                                                        EDGAR L. DE CASTRO                                                                                                                                  PAGE 1
                                                                                                                                                                                                                                                      ..
..
                                                                                                                                                                                                                                                                "


CLASSIFICATIONS OF DT PROBLEMS:

1. Single Stage Decision Problems
      A decision is m~de only once.                                                               ~"

                                                                                                                                                                                                    ~




2. Multiple Stage/Sequential Decision Problems
     Decisions are made one after another.
                                                                                                                                                                                                                                                      ..
                                                                                                                                                                                                                                                                f:
3. Discrete DT Problems
      The alternative courses of actions and states of nature are finite.
                                                                                                                                                                                                                  ./
4. Continuous DT Problems
     The alternative courses of actions and states of nature are infinite.

DT Problems can also be classified as those with or without
experimentation. Experimentation is perfoaned to obtain additional
information that will aid the decision maker.

I. DISCRETE DECISION THEORY PROBLEMS
DECISION TREES

     A discrete DT problem can be represented pictorially using a tree.
diagram or decision tree. It chronologically depicts the sequence of
actions and events as they unfold.

      A square node ( D) precedes the set of possible actions that can
be taken by the decision maker. A round node (0 ) precedes the set of
events or states of nature that could be encountered after a decision is
made. The nodes are connectedby branches. (""- )



111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111


DECISION THEORY                                                                                        EDGAR L. DE CASTRO                                                                                                                              PAGE 2



                                                                                                                                            :.:.~.;.
                                                                                                                                                 .'                " '.
'.

                                                                                                                                                                                                                                                                                                                   '. .


                   EXAMPLE:                                                                                                                                                                                                                                                                                        i.
                                                                                                                                                                                                                                                                                                                   ,

                                                                                                                                                                                                                                                                                                                   I




                                                                                                                                                                                                                                                             . ..




                   DECISIONS UNDER RISK AND UNCERTAINTY

                   Consider a DT problem with m alternative courses of actions and a
                   maximum of n events or states of nature for each alternativecourse of
                   action.

                   Defme:                                          Ai          = alternative course of action i; i = 1, 2, . . .,m

                                                                    q)j=state of nature j; j = 1, 2, . . .,n

                   The decision matrix of payoffs is given by :

                                                                                                      q)l                                                         q)2                                          ...                                                          q)11
                                              Al                                        v(A1,q)1)                                             veAl ,q)2)                                                       ...                                          v(Abn)
                                              A2                                        v(A2'1)                                               v(A2'2)                                                          ...                                          v(A2 '11)
                                                 .                                                        .                                                              .                                         .                                                         .
                                                 .                                                        .                                                              .                                        .                                                          .
                                                 .                                                        .                                                              .                                        .                                                          .
                                              Am                                          v(Am '1)                                               V(Am'2)                                                       ...                                         v(An"l1 )

                   I"   11/1"   III"   III"   111111111111"   11/1111111111111111111111111111111"      111111111111111/1/11111111111111111111111/11/11111111111111111"       111/111111111""   II"   1111 ""   lilli/II   n"   111111 n n I"   nnn   111111 nn"   1I11 ""   III!   11111"   11111/1111111111111


                   DECISION THEORY                                                                                   EDGAR L. DE CASTRO                                                                                                                           PAGE 3



.
    .," .....
      ..........
                                                                                                    .' ...
                                                                                                     . .
                                                                                                      .
                                                                                                             " '. ~
                                                                                                               . ..
                                                                                                           .. '. .
                                                                                                                   '.'.
                                                                                                                                                                                                                                                                            .','.                          ,',',
                                                                                                                                                                                                                                                                                                         ..' .
                                                                                                                                                                                                                                                                                                           .:.
.~
                                                                                                                                                                                                                                                                               ll1
                                                                                                                                                                                                                                                                                 ."j'
                                                                                                                                                                                                                                                                                  "
 A. LAPLACE CRITERION

      This criterion is based on what is mown as the principle of
 insufficientreason. Her~the probabilitiesassociatedwith the'occurrence
 of the event                                                                     is uriknown.We do not have,Su(.ticie~~,leason
                                                                                                                              to
                                                                                                                                                                                                                                                                            ',.


                                                                 rjJj                                                                                                                                                                                 ,           "



 conclude that the probabilities are different. Hence we 'assume that all                                                                                                                                                                                                  "

 events are equally likely, i.e.                                                                                                                                                                                                                                            ....
                                                                                                                      .                                                  1                                                                                                 f.. .
                                                                                                               P( t/J = rjJ . ) = -                                                                                                                                        ~.:-.
                                                                                                                                                  }                     n
                                                                                                                                                                                                                                                                           ;-:..
                                                                                                                                                                                                                                                            ...



 Then, the optimal decision rule is to select action                                                                                                                                                                     '51t           corresponding to                   "
                                                                                                                                                                                                                                                                           ....


                                                                                                                                                                                                                                                                           ,,'




B. MINIMAX (MAXIMIN) CRITERION

      This is the most conservative criterion since it is based on making
the best out of the worst possible conditions. For each possible decision
alternative, we select the worst condition and then select the alternative
corresponding to the best of the worst conditions.

The MINIMAX strategy is given by:
                                                                                                                min
                                                                                                                     A;          f max{V(Ai,rjJ
                                                                                                                                 L. r/J.
                                                                                                                                      J
                                                                                                                                                                                           j }
                                                                                                                                                                                                       ]
The MAXIMIN strategy is given by:
                                                                                                                max
                                                                                                                      A;           [min {v( Ai , rjJ ]
                                                                                                                                      j     r/J
                                                                                                                                                   j                                             }




111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111


DECISION THEORY                                                                                        EDGAR L. DE CASTRO                                                                                                                                         PAGE 4


                                                                                      . : .: .'. :'. ".: ,';                                      ~::..~.' '.                                 . ...,       :...:..: :         ~.:r::...           ,'.'.:.
C. SAVAGE MINIMAX REGRET CRITERION

      The MINIMAX rule is an extremely conservative type of decision
rule. The savage MINIMAx regret criterion assUmes that a new .loss                                                                                                                                                                                                           ,.
                                                                                                                                                                                                                                                                             I
                                                                                                                                                                                                                                                                             r
                                                                                                                                                                                                                                                                             ,.
matrix is constructed id which v(Ai, (Jj) is replaced by r(A~,..(J)' which
                                                                 j                                                                                                                                                                                                           1.

is defined by:
                                                                                                                                                                                                                                                      ..

                                                                                             max { v( Ak , (Jj                                             )}        -       v( Ai , (Jj ),                                      if v is profit
                                                                                              Ak
                                                                                                                                                                                                                                                           ,                 r
                                                                                                                                                                                                                                                                             t.
                                                                                                  v(Ai,(J j) - min{v(Ak,(J )}, if v i~loss
                                                                                                                         j                                                                                                                                 II


                                                                                                                                                       Ak                                                            .,/

                                                                                                                                                                                                                    .                  .
Once the loss matrix is constructed using the above fonnula, we can
now apply the MINIMAX criterion defined in b.

D. HURWICZ CRITERION

                        This criterion represents a range of attitudes from the most
optimistic to the most pessimistic.

Under the most optimistic conditions, one would choose the action
yielding:
                                                                                                               max max {v( Ai , (Jj }
                                                                                                                Ai { I               }       t/>




Under the most pessimistic conditions, the chosen action cOlTesponds
to:                                                                                                                                                                                                          .




                                                                                                               max                 min {v( Ai , (Jj }
                                                                                                                     A.             A..
                                                                                                                      t           { '1')             }




111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111



DECISIONTHEORY                                                                                         EDGARL. DE CASTRO                                                                                                                              PAGE5


                                                                                                                                                   " '..                                ".     .,     :     '..: ;.," .:j"......                                . .   . ..
The Hurwicz criterionstrikes a balancebetween extremepessimism and
            extreme optimism by weighing the above conditions by respective                                                                                                                                                                                       I

            weights a and (1- a), where 0 < a <1. That is the action selected is                                                                                                                                                                                  t
                                                                                                                                                                                                                                                                  ,


            that which yields:'                           .
                                                                                                                                                                                                                     .,
                                                        ma..'Xa m~ v( Ai , r/J ) + (1 - a )min v( Ai , r/J )
                                                                             j                           j
                                                            Ai        {              t/J   j                                                                     t/J
                                                                                                                                                                   j                                      }
                                                                                                                                                                                                                    ..

            [Note the above formulas represent the case where payoffs are expressed
            as profits]

            If a = 1, the decision rule is referred to as the MAXIMAX RULE, and if
            a = 0, the decision rule becomes the MAXIMIN RULE. For the case
            where the payoff represent costs, the decision rule is given by:
                                                         min a min v( Ai , f/Jj ) + (1- a) max v( Ai , f/J )
                                                                                                         j
                                                            ~ {                     ~                                                                           ~                                        }

            E. BAYES' RULE

                              Here, weasswne that the probabilities associated with each state
            of natureareknown.Let                                                                                                                                                                   .      .


                                                                                                          P{ljJ           = ljJj} = Pj
            The action which minimizes (maximizes) the cost (profit) is selected.
            This is given by:




            The backward induction approach is used. With the aid of a decision
            tree, expected values are computed each time a round node is
            encountered and the above decision rule is utilized each time a square
            node is encountered ,i.e., a decision is made each time a square node is
            encountered.
            IIIIIIIIIIIII!IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111

            DECISIONTHEORY                                                                     EDGARL. DE CASTRO                                                                                                    PAGE6


                                                                                  . . . .: .:
                                                                                    . .~ .
.,". ':".                   .,',,1, ..
                             . ','.                                                                    i.L                   ,.
..
                                                                                                                                                                                                                                                                 I .'
                                                                                                                                                                                                                                                                 t,




               F. EXPECTED VALUE-VARIANCE CRITERION

                               This is an extension of the expected value criterion. Here we
               simultaneously maximi~e profit and minimize the variance of the profit.                                                                                                                                                                           .   .

               If Z represents profit as a random variable with variance q:, then the
               criterion is given by:

                                                                                           maximize E(z)                                       - Kvar(z)
           where K is any specified constant. If Z represents cost:                                                                                                                                                                                              .' .
                                                                                                                                                                                  ./

                                                                                       minimize E(z) + Kvar(z)

               G. DECISION MAKING WITH EXPERIMENTATION

                In some situations, it may be viable to secure additional
           information tp revise the original estimates of the probability of
           occurrence of the state of nature.

           DEFINITION 1.2 :

           Preposterior Analysis considers the question of deciding whether or not
           it would be worthwhile to get additional information or to perfonn
           further experimentation.

           DEFINITION 1.3 :

           Posterior Analysis deals with the optimal choice and evaluation of an
           action subsequent to all experimentation and testing using the
           experimental results.




           /1/1/1111111111111111111111/111111111111111111111111111111111111111111111/111111/11111111/1111111111111111111/111111111111111111/1/11111/1111111111111111111111111111111111111111111111111/1/111111111/1/111111111111111111111/1/111111/11111111111

           DECISION THEORY                                                                 EDGAR L. DE CASTRO                                                                                                          PAGE 7



                                                                                                                              "
".':.:::: >.         "          :.-:'.:.:..
---.......-----




DEFINITION 1.4 :

Prior probabilities are the initial probabilities assumed without the
benefit of experiment~tion. Posterior probabilities refer to the revised
probability values obtamed after experimentation.           . -




Let: Pj                                =prior                             probability estimate of event (Jj
                          P{Zkl(Jj}  = conditional probability of experimental                                                                                                                                                                        (jutcome Zk
                          P{(Jj IZk} = posterior probability of event (Jj

The experimental results are assumed to be given by Zk, k = 1, 2, ... 1.
The conditional probability can be considered to be a measure of the
reliability of the ~xperiment. The idea is to calculate the posterior
probabilities by combining the prior probabilities and the conditional
probabilities of experimental outcome Zk. The posterior probabilities are
given by:


                                                                                                                                                                                    m
                                                                                                                                                                                  L P{Zk l(Ji}P{(Ji}
                                                                                                                                                                                  i=1



Once the posterior probabilities are calculated, the original problem can
be viewed as a multiple stage/sequential DT problem. The first stage
involves the decision of whether to perform additional experimentation
or not. Once this is decided, the outcomes of the experiment are
considered together with the original set of decision alternatives and
events.




1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/1111111111111111111111111111111111111111111111111111111111111111


DECISIONTHEORY                                                                                         EDGARL. DE CASTRO                                                                                                                              PAGE8


                                                                                            . '.            .
DEFINITION 1.5 :

              A perfect infonnation source would provide, with 100% reliability,
              which of the states of nature wouldoccur.    .
                                                                                                                    >'.



              Define:                                         EPPI = expected profit from a perfect information source
                                                              EVPI = expected value of the perfect infonnation source
                                                              EP = Bayes' expected profit without experimentation

              Then:
                                                                                                                                                                                                                                   .,
                                                              EVPI = EPPI                                                 - EP
              where:

                                                                                                       11


                                                               E VPI                       = L Pj                            * max { v( Ai , fjJ ) }
                                                                                                                                               j
                                                                                                 . j=1                                   Ai




              EVPI is easily seen as a measure of the maximum amount a decision
              maker should be willing to pay for additional infonnation.

              Define:                                         EVSI = expected value of sample information
                                                              ENOS = expected net gain from sampling
                                                              CAI = cost of getting additional information




              111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111



              DECISIONTHEORY                                                                                         EDGARL. DE CASTRO                                                                                                                              PAGE9


     . '.' '.                             ..                                                                                                                                                                                                                                ....
    . .' .'                                                                                                                                                     .,
.                                 ..' ',' I"'.               ,"
            " ..                   ..
,
                                                                        .,
                                                                         ,

          Then:                                                                                                                                                                                                                                               I
                                                                                                                                                                                                                                                              , ,




                                                                                                                                                                                                                                                              "
                                                                                                                                                                                                      ,. -
                                                                                                                                                                                                         ,
                                             ENGS                  = EVSI                    - CAI
                                                                                                                                                                                                         ..


          The information source would be viable if ENGS > O.

          II. CONTINUOUS DECISION THEORY.

                As previously mentioned, continuous decision theory problems
          refer to those where the number of alternatives and/or states of nature
          can be considered infmite. The optimization model in this case is given
          by:                                                                .



                                                                                 max f(A)                   =J: v(A,t/J)htjJ(t/J)dt/J
          where:


          htjJ(t/J)            = prior                  distribution function of the states of nature

                In the above model, it is assumed that no additional infonnation is
          available and the expectation is evaluated with respect to the prior
          distribution of the states of nature. If additional infonnation is available,
          we update the prior distribution of the states of nature by detennining its
          posterior distribution, which is nothing but the conditional distribution
          of the states of nature given the experimental outcome. Hence, the
          optimization converts to:



          IIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111

          DECISIONTHEORY                                                               EDGARL. DE CASTRO                                                                                              PAGE 10


. .....                  ..:...:....                                             ...                                                              .      ;...:..' :.. ....
                                                                                                                                                                    .                                               .          . . ..
-------




                                                                                  .    max f(A) =                                              f:              v(A,f/J)ht/>IZ=z (f/J)df/J


      where:
                                                                                                         '..

.jI
 I     hfj)IZ=z (rjJ)                             = conditional distribution of the state of nature' given the
                                                          experimental outcome
 :!
 I
 I
..J




       hZIfj)(z)                       = conditional distribution of the experimental outcome given the
                                                 state of nature

       hz (z)                    = marginal distribution function of the experimental outcomes

      where:



      LEIBNIZ' RULE

                              LEIBNIZ' Rule is applied to find the derivative'of a function
      which contains integrals. Consider a function in one variable A:

                                                       d              b ig
                                                                       b               db                                                                                                                                                                   da
                                                   - fa g(A,rjJ)drjJ=fa ~rjJ+g(A,b)--g(A,a)-
                                                   ciA                  8A             ciA                                                                                                                                                                  ciA




      111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111


      DECISION THEORY                                                                                          EDGAR L. DE CASTRO                                                                                                                           PAGE 11



                                                                                          '. '.: '. . :

More Related Content

What's hot

Decision making environment
Decision making environmentDecision making environment
Decision making environmentshubhamvaghela
 
Simple regression and correlation
Simple regression and correlationSimple regression and correlation
Simple regression and correlationMary Grace
 
Cost of Preference Capital Soved Problems-kp
Cost of Preference Capital Soved Problems-kpCost of Preference Capital Soved Problems-kp
Cost of Preference Capital Soved Problems-kpuma reur
 
Leverage (Operating, financial & combined leverage)
Leverage (Operating, financial & combined leverage)Leverage (Operating, financial & combined leverage)
Leverage (Operating, financial & combined leverage)Yamini Kahaliya
 
Internal Environmental Analysis
Internal Environmental AnalysisInternal Environmental Analysis
Internal Environmental AnalysisFaiz Alwi
 
Decision theory
Decision theoryDecision theory
Decision theorySurekha98
 
Dominance method
Dominance methodDominance method
Dominance methodpooja rani
 
Cost volume profit analysis.
Cost volume profit analysis.Cost volume profit analysis.
Cost volume profit analysis.Varadraj Bapat
 
Perception in OB
Perception in OBPerception in OB
Perception in OBRaghav Jha
 
decision making criterion
decision making criteriondecision making criterion
decision making criterionGaurav Sonkar
 
Risk and Return Analysis
Risk and Return AnalysisRisk and Return Analysis
Risk and Return AnalysisRamziya Begam
 

What's hot (20)

probable-error.pdf
probable-error.pdfprobable-error.pdf
probable-error.pdf
 
Decision making environment
Decision making environmentDecision making environment
Decision making environment
 
Game theory
Game theoryGame theory
Game theory
 
Simple regression and correlation
Simple regression and correlationSimple regression and correlation
Simple regression and correlation
 
Linear programing
Linear programingLinear programing
Linear programing
 
Cost of Preference Capital Soved Problems-kp
Cost of Preference Capital Soved Problems-kpCost of Preference Capital Soved Problems-kp
Cost of Preference Capital Soved Problems-kp
 
capital_budgeting
capital_budgetingcapital_budgeting
capital_budgeting
 
Leverage (Operating, financial & combined leverage)
Leverage (Operating, financial & combined leverage)Leverage (Operating, financial & combined leverage)
Leverage (Operating, financial & combined leverage)
 
Internal Environmental Analysis
Internal Environmental AnalysisInternal Environmental Analysis
Internal Environmental Analysis
 
Decision theory
Decision theoryDecision theory
Decision theory
 
Dominance method
Dominance methodDominance method
Dominance method
 
Cost volume profit analysis.
Cost volume profit analysis.Cost volume profit analysis.
Cost volume profit analysis.
 
Linear programming ppt
Linear programming pptLinear programming ppt
Linear programming ppt
 
Perception in OB
Perception in OBPerception in OB
Perception in OB
 
decision making criterion
decision making criteriondecision making criterion
decision making criterion
 
Risk and Return Analysis
Risk and Return AnalysisRisk and Return Analysis
Risk and Return Analysis
 
An Introduction To Business Strategy
An Introduction To Business StrategyAn Introduction To Business Strategy
An Introduction To Business Strategy
 
Strategy Formulation and Implementation
Strategy Formulation and ImplementationStrategy Formulation and Implementation
Strategy Formulation and Implementation
 
Quantitative Techniques
Quantitative TechniquesQuantitative Techniques
Quantitative Techniques
 
Game theory
Game theoryGame theory
Game theory
 

Viewers also liked

DECISION THEORY WITH EXAMPLE
DECISION THEORY WITH EXAMPLEDECISION THEORY WITH EXAMPLE
DECISION THEORY WITH EXAMPLEAnasuya Barik
 
Decision theory approach in management
Decision theory approach in managementDecision theory approach in management
Decision theory approach in managementsanna1
 
Decision making ppt
Decision making pptDecision making ppt
Decision making pptashgrover
 
Inroduction to Decision Theory and Decision Making Under Certainty
Inroduction to Decision Theory and Decision Making Under CertaintyInroduction to Decision Theory and Decision Making Under Certainty
Inroduction to Decision Theory and Decision Making Under CertaintyAbhi23396
 
Decision Making Process
Decision Making ProcessDecision Making Process
Decision Making ProcessAima Masood
 
DECISION MAKING POWERPOINT
DECISION MAKING POWERPOINT DECISION MAKING POWERPOINT
DECISION MAKING POWERPOINT Andrew Schwartz
 
Relationship between IRR & NPV
Relationship between IRR & NPVRelationship between IRR & NPV
Relationship between IRR & NPVZeeshan Ali
 
Npv and IRR, a link to Project Management
Npv and IRR, a link to Project ManagementNpv and IRR, a link to Project Management
Npv and IRR, a link to Project ManagementUjjwal Joshi
 
Quantitative techniques introduction 19 pages
Quantitative techniques introduction 19 pagesQuantitative techniques introduction 19 pages
Quantitative techniques introduction 19 pagestaniyakhurana
 
Decision making & problem solving
Decision making & problem solvingDecision making & problem solving
Decision making & problem solvingashish1afmi
 
Decision tree powerpoint presentation templates
Decision tree powerpoint presentation templatesDecision tree powerpoint presentation templates
Decision tree powerpoint presentation templatesSlideTeam.net
 
Decision Analysis
Decision AnalysisDecision Analysis
Decision Analysiss junaid
 

Viewers also liked (20)

Ppt on decision theory
Ppt on decision theoryPpt on decision theory
Ppt on decision theory
 
DECISION THEORY WITH EXAMPLE
DECISION THEORY WITH EXAMPLEDECISION THEORY WITH EXAMPLE
DECISION THEORY WITH EXAMPLE
 
Decision Theory
Decision TheoryDecision Theory
Decision Theory
 
Decision theory approach in management
Decision theory approach in managementDecision theory approach in management
Decision theory approach in management
 
Decision theory Problems
Decision theory ProblemsDecision theory Problems
Decision theory Problems
 
Decision theory
Decision theoryDecision theory
Decision theory
 
Decision theory
Decision theoryDecision theory
Decision theory
 
Decision making ppt
Decision making pptDecision making ppt
Decision making ppt
 
Inroduction to Decision Theory and Decision Making Under Certainty
Inroduction to Decision Theory and Decision Making Under CertaintyInroduction to Decision Theory and Decision Making Under Certainty
Inroduction to Decision Theory and Decision Making Under Certainty
 
Decision Making Process
Decision Making ProcessDecision Making Process
Decision Making Process
 
Decision making
Decision makingDecision making
Decision making
 
DECISION MAKING POWERPOINT
DECISION MAKING POWERPOINT DECISION MAKING POWERPOINT
DECISION MAKING POWERPOINT
 
NPV v/s IRR
NPV v/s IRRNPV v/s IRR
NPV v/s IRR
 
Relationship between IRR & NPV
Relationship between IRR & NPVRelationship between IRR & NPV
Relationship between IRR & NPV
 
Npv and IRR, a link to Project Management
Npv and IRR, a link to Project ManagementNpv and IRR, a link to Project Management
Npv and IRR, a link to Project Management
 
Quantitative techniques introduction 19 pages
Quantitative techniques introduction 19 pagesQuantitative techniques introduction 19 pages
Quantitative techniques introduction 19 pages
 
Decision making
Decision makingDecision making
Decision making
 
Decision making & problem solving
Decision making & problem solvingDecision making & problem solving
Decision making & problem solving
 
Decision tree powerpoint presentation templates
Decision tree powerpoint presentation templatesDecision tree powerpoint presentation templates
Decision tree powerpoint presentation templates
 
Decision Analysis
Decision AnalysisDecision Analysis
Decision Analysis
 

Similar to Decision theory

Clear Channel Outdoor Media Planning Guide
Clear Channel Outdoor Media Planning GuideClear Channel Outdoor Media Planning Guide
Clear Channel Outdoor Media Planning GuideRomulus Stoian
 
Java Apis For Imaging Enterprise-Scale, Distributed 2d Applications
Java Apis For Imaging Enterprise-Scale, Distributed 2d ApplicationsJava Apis For Imaging Enterprise-Scale, Distributed 2d Applications
Java Apis For Imaging Enterprise-Scale, Distributed 2d Applicationswhite paper
 
Resident Individual Income Tax Return - EZ
Resident Individual Income Tax Return - EZResident Individual Income Tax Return - EZ
Resident Individual Income Tax Return - EZtaxman taxman
 
Reforestation Tax Credit
Reforestation Tax CreditReforestation Tax Credit
Reforestation Tax Credittaxman taxman
 
Intra Africa Fibre Fiber Map
Intra Africa Fibre Fiber MapIntra Africa Fibre Fiber Map
Intra Africa Fibre Fiber MapEd Dodds
 
Underestimate of Tax
Underestimate of TaxUnderestimate of Tax
Underestimate of Taxtaxman taxman
 
Designnet > 09/11
Designnet > 09/11Designnet > 09/11
Designnet > 09/11guest787af7
 
Demar Brochure Final Jan 2012
Demar Brochure Final Jan 2012Demar Brochure Final Jan 2012
Demar Brochure Final Jan 2012ImyMartinez
 
Cv S.Uiterwijk
Cv S.UiterwijkCv S.Uiterwijk
Cv S.UiterwijkSuiterwijk
 
Illinois Launch Presentation
Illinois Launch PresentationIllinois Launch Presentation
Illinois Launch PresentationEDeLaPaz_PDA
 
Study states-background
Study states-backgroundStudy states-background
Study states-backgroundRahul Bhargava
 
Evaluasi nilai berkala i gasal_kpn_12-13
Evaluasi nilai berkala i gasal_kpn_12-13Evaluasi nilai berkala i gasal_kpn_12-13
Evaluasi nilai berkala i gasal_kpn_12-13Didik Purwiyanto Vay
 
Studyforprogrammer
StudyforprogrammerStudyforprogrammer
Studyforprogrammerguest0a0b14
 
before traveling
before travelingbefore traveling
before travelingJune Song
 
BA Champs PowerPoint AGM Presentation
BA Champs PowerPoint AGM PresentationBA Champs PowerPoint AGM Presentation
BA Champs PowerPoint AGM PresentationEugene Cheng
 

Similar to Decision theory (20)

Clear Channel Outdoor Media Planning Guide
Clear Channel Outdoor Media Planning GuideClear Channel Outdoor Media Planning Guide
Clear Channel Outdoor Media Planning Guide
 
Java Apis For Imaging Enterprise-Scale, Distributed 2d Applications
Java Apis For Imaging Enterprise-Scale, Distributed 2d ApplicationsJava Apis For Imaging Enterprise-Scale, Distributed 2d Applications
Java Apis For Imaging Enterprise-Scale, Distributed 2d Applications
 
Resident Individual Income Tax Return - EZ
Resident Individual Income Tax Return - EZResident Individual Income Tax Return - EZ
Resident Individual Income Tax Return - EZ
 
Reforestation Tax Credit
Reforestation Tax CreditReforestation Tax Credit
Reforestation Tax Credit
 
Intra Africa Fibre Fiber Map
Intra Africa Fibre Fiber MapIntra Africa Fibre Fiber Map
Intra Africa Fibre Fiber Map
 
Underestimate of Tax
Underestimate of TaxUnderestimate of Tax
Underestimate of Tax
 
P13 033
P13 033P13 033
P13 033
 
Lecture 5
Lecture 5Lecture 5
Lecture 5
 
P523
P523P523
P523
 
Designnet 09/11
Designnet 09/11Designnet 09/11
Designnet 09/11
 
Designnet > 09/11
Designnet > 09/11Designnet > 09/11
Designnet > 09/11
 
Demar Brochure Final Jan 2012
Demar Brochure Final Jan 2012Demar Brochure Final Jan 2012
Demar Brochure Final Jan 2012
 
Cv S.Uiterwijk
Cv S.UiterwijkCv S.Uiterwijk
Cv S.Uiterwijk
 
Illinois Launch Presentation
Illinois Launch PresentationIllinois Launch Presentation
Illinois Launch Presentation
 
Study states-background
Study states-backgroundStudy states-background
Study states-background
 
Evaluasi nilai berkala i gasal_kpn_12-13
Evaluasi nilai berkala i gasal_kpn_12-13Evaluasi nilai berkala i gasal_kpn_12-13
Evaluasi nilai berkala i gasal_kpn_12-13
 
P13 021
P13 021P13 021
P13 021
 
Studyforprogrammer
StudyforprogrammerStudyforprogrammer
Studyforprogrammer
 
before traveling
before travelingbefore traveling
before traveling
 
BA Champs PowerPoint AGM Presentation
BA Champs PowerPoint AGM PresentationBA Champs PowerPoint AGM Presentation
BA Champs PowerPoint AGM Presentation
 

More from De La Salle University-Manila

Chapter3 general principles of discrete event simulation
Chapter3   general principles of discrete event simulationChapter3   general principles of discrete event simulation
Chapter3 general principles of discrete event simulationDe La Salle University-Manila
 

More from De La Salle University-Manila (20)

Queueing theory
Queueing theoryQueueing theory
Queueing theory
 
Queueing theory
Queueing theoryQueueing theory
Queueing theory
 
Queuing problems
Queuing problemsQueuing problems
Queuing problems
 
Verfication and validation of simulation models
Verfication and validation of simulation modelsVerfication and validation of simulation models
Verfication and validation of simulation models
 
Markov exercises
Markov exercisesMarkov exercises
Markov exercises
 
Markov theory
Markov theoryMarkov theory
Markov theory
 
Game theory problem set
Game theory problem setGame theory problem set
Game theory problem set
 
Decision theory handouts
Decision theory handoutsDecision theory handouts
Decision theory handouts
 
Sequential decisionmaking
Sequential decisionmakingSequential decisionmaking
Sequential decisionmaking
 
Decision theory blockwood
Decision theory blockwoodDecision theory blockwood
Decision theory blockwood
 
Random variate generation
Random variate generationRandom variate generation
Random variate generation
 
Random number generation
Random number generationRandom number generation
Random number generation
 
Monte carlo simulation
Monte carlo simulationMonte carlo simulation
Monte carlo simulation
 
Input modeling
Input modelingInput modeling
Input modeling
 
Conceptual modeling
Conceptual modelingConceptual modeling
Conceptual modeling
 
Chapter3 general principles of discrete event simulation
Chapter3   general principles of discrete event simulationChapter3   general principles of discrete event simulation
Chapter3 general principles of discrete event simulation
 
Comparison and evaluation of alternative designs
Comparison and evaluation of alternative designsComparison and evaluation of alternative designs
Comparison and evaluation of alternative designs
 
Chapter2
Chapter2Chapter2
Chapter2
 
Chapter1
Chapter1Chapter1
Chapter1
 
Variance reduction techniques (vrt)
Variance reduction techniques (vrt)Variance reduction techniques (vrt)
Variance reduction techniques (vrt)
 

Recently uploaded

INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptx
INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptxINTRODUCTION TO CATHOLIC CHRISTOLOGY.pptx
INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptxHumphrey A Beña
 
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdf
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdfInclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdf
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdfTechSoup
 
What is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPWhat is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPCeline George
 
Choosing the Right CBSE School A Comprehensive Guide for Parents
Choosing the Right CBSE School A Comprehensive Guide for ParentsChoosing the Right CBSE School A Comprehensive Guide for Parents
Choosing the Right CBSE School A Comprehensive Guide for Parentsnavabharathschool99
 
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITY
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITYISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITY
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITYKayeClaireEstoconing
 
Earth Day Presentation wow hello nice great
Earth Day Presentation wow hello nice greatEarth Day Presentation wow hello nice great
Earth Day Presentation wow hello nice greatYousafMalik24
 
Science 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxScience 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxMaryGraceBautista27
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTiammrhaywood
 
Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Jisc
 
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxiammrhaywood
 
DATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersDATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersSabitha Banu
 
Judging the Relevance and worth of ideas part 2.pptx
Judging the Relevance  and worth of ideas part 2.pptxJudging the Relevance  and worth of ideas part 2.pptx
Judging the Relevance and worth of ideas part 2.pptxSherlyMaeNeri
 
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONTHEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONHumphrey A Beña
 
Barangay Council for the Protection of Children (BCPC) Orientation.pptx
Barangay Council for the Protection of Children (BCPC) Orientation.pptxBarangay Council for the Protection of Children (BCPC) Orientation.pptx
Barangay Council for the Protection of Children (BCPC) Orientation.pptxCarlos105
 
Keynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designKeynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designMIPLM
 
ENGLISH6-Q4-W3.pptxqurter our high choom
ENGLISH6-Q4-W3.pptxqurter our high choomENGLISH6-Q4-W3.pptxqurter our high choom
ENGLISH6-Q4-W3.pptxqurter our high choomnelietumpap1
 
Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Celine George
 
Grade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxGrade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxChelloAnnAsuncion2
 

Recently uploaded (20)

FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptxFINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
 
INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptx
INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptxINTRODUCTION TO CATHOLIC CHRISTOLOGY.pptx
INTRODUCTION TO CATHOLIC CHRISTOLOGY.pptx
 
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdf
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdfInclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdf
Inclusivity Essentials_ Creating Accessible Websites for Nonprofits .pdf
 
What is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERPWhat is Model Inheritance in Odoo 17 ERP
What is Model Inheritance in Odoo 17 ERP
 
Choosing the Right CBSE School A Comprehensive Guide for Parents
Choosing the Right CBSE School A Comprehensive Guide for ParentsChoosing the Right CBSE School A Comprehensive Guide for Parents
Choosing the Right CBSE School A Comprehensive Guide for Parents
 
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITY
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITYISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITY
ISYU TUNGKOL SA SEKSWLADIDA (ISSUE ABOUT SEXUALITY
 
Earth Day Presentation wow hello nice great
Earth Day Presentation wow hello nice greatEarth Day Presentation wow hello nice great
Earth Day Presentation wow hello nice great
 
Raw materials used in Herbal Cosmetics.pptx
Raw materials used in Herbal Cosmetics.pptxRaw materials used in Herbal Cosmetics.pptx
Raw materials used in Herbal Cosmetics.pptx
 
Science 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxScience 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptx
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
 
Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...
 
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptxECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
ECONOMIC CONTEXT - PAPER 1 Q3: NEWSPAPERS.pptx
 
DATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersDATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginners
 
Judging the Relevance and worth of ideas part 2.pptx
Judging the Relevance  and worth of ideas part 2.pptxJudging the Relevance  and worth of ideas part 2.pptx
Judging the Relevance and worth of ideas part 2.pptx
 
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONTHEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
 
Barangay Council for the Protection of Children (BCPC) Orientation.pptx
Barangay Council for the Protection of Children (BCPC) Orientation.pptxBarangay Council for the Protection of Children (BCPC) Orientation.pptx
Barangay Council for the Protection of Children (BCPC) Orientation.pptx
 
Keynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-designKeynote by Prof. Wurzer at Nordex about IP-design
Keynote by Prof. Wurzer at Nordex about IP-design
 
ENGLISH6-Q4-W3.pptxqurter our high choom
ENGLISH6-Q4-W3.pptxqurter our high choomENGLISH6-Q4-W3.pptxqurter our high choom
ENGLISH6-Q4-W3.pptxqurter our high choom
 
Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17
 
Grade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxGrade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptx
 

Decision theory

  • 1. I . I . ! I j ! . ! . f : DECISION THEORY DEFINITION 1.1 : DECISION THEORY (DT) is a set of concepts, principles, tools and techniques that aid the decision maker in dealing with compl~x decision problems under uncertainty. .- COMPONENTS OF A DT PROBLEM: 1. THE DECISION MAKER 2. ALTERNATIVE COURSES OF ACTION This is the controllable aspect of the problem. 3. STATES OF NATURE OR EVENTS These are the scenarios or states of the environmen.tnot under the control of the decision maker. The events defmed should be mutually exclusive and collectively exhaustive. 4. CONSEQUENCES The consequences that must be assessed by the decision maker are measures of the net bertefit, payoff, cost or revenue received by the decision maker. There is a consequence (or vector of consequences) associated with each action-event pair. The consequences resutilmarized a decisionmatrix. a in . 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111II DECISION THEORY EDGAR L. DE CASTRO PAGE 1 ..
  • 2. .. " CLASSIFICATIONS OF DT PROBLEMS: 1. Single Stage Decision Problems A decision is m~de only once. ~" ~ 2. Multiple Stage/Sequential Decision Problems Decisions are made one after another. .. f: 3. Discrete DT Problems The alternative courses of actions and states of nature are finite. ./ 4. Continuous DT Problems The alternative courses of actions and states of nature are infinite. DT Problems can also be classified as those with or without experimentation. Experimentation is perfoaned to obtain additional information that will aid the decision maker. I. DISCRETE DECISION THEORY PROBLEMS DECISION TREES A discrete DT problem can be represented pictorially using a tree. diagram or decision tree. It chronologically depicts the sequence of actions and events as they unfold. A square node ( D) precedes the set of possible actions that can be taken by the decision maker. A round node (0 ) precedes the set of events or states of nature that could be encountered after a decision is made. The nodes are connectedby branches. (""- ) 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISION THEORY EDGAR L. DE CASTRO PAGE 2 :.:.~.;. .' " '.
  • 3. '. '. . EXAMPLE: i. , I . .. DECISIONS UNDER RISK AND UNCERTAINTY Consider a DT problem with m alternative courses of actions and a maximum of n events or states of nature for each alternativecourse of action. Defme: Ai = alternative course of action i; i = 1, 2, . . .,m q)j=state of nature j; j = 1, 2, . . .,n The decision matrix of payoffs is given by : q)l q)2 ... q)11 Al v(A1,q)1) veAl ,q)2) ... v(Abn) A2 v(A2'1) v(A2'2) ... v(A2 '11) . . . . . . . . . . . . . . . Am v(Am '1) V(Am'2) ... v(An"l1 ) I" 11/1" III" III" 111111111111" 11/1111111111111111111111111111111" 111111111111111/1/11111111111111111111111/11/11111111111111111" 111/111111111"" II" 1111 "" lilli/II n" 111111 n n I" nnn 111111 nn" 1I11 "" III! 11111" 11111/1111111111111 DECISION THEORY EDGAR L. DE CASTRO PAGE 3 . .," ..... .......... .' ... . . . " '. ~ . .. .. '. . '.'. .','. ,',', ..' . .:.
  • 4. .~ ll1 ."j' " A. LAPLACE CRITERION This criterion is based on what is mown as the principle of insufficientreason. Her~the probabilitiesassociatedwith the'occurrence of the event is uriknown.We do not have,Su(.ticie~~,leason to ',. rjJj , " conclude that the probabilities are different. Hence we 'assume that all " events are equally likely, i.e. .... . 1 f.. . P( t/J = rjJ . ) = - ~.:-. } n ;-:.. ... Then, the optimal decision rule is to select action '51t corresponding to " .... ,,' B. MINIMAX (MAXIMIN) CRITERION This is the most conservative criterion since it is based on making the best out of the worst possible conditions. For each possible decision alternative, we select the worst condition and then select the alternative corresponding to the best of the worst conditions. The MINIMAX strategy is given by: min A; f max{V(Ai,rjJ L. r/J. J j } ] The MAXIMIN strategy is given by: max A; [min {v( Ai , rjJ ] j r/J j } 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISION THEORY EDGAR L. DE CASTRO PAGE 4 . : .: .'. :'. ".: ,'; ~::..~.' '. . ..., :...:..: : ~.:r::... ,'.'.:.
  • 5. C. SAVAGE MINIMAX REGRET CRITERION The MINIMAX rule is an extremely conservative type of decision rule. The savage MINIMAx regret criterion assUmes that a new .loss ,. I r ,. matrix is constructed id which v(Ai, (Jj) is replaced by r(A~,..(J)' which j 1. is defined by: .. max { v( Ak , (Jj )} - v( Ai , (Jj ), if v is profit Ak , r t. v(Ai,(J j) - min{v(Ak,(J )}, if v i~loss j II Ak .,/ . . Once the loss matrix is constructed using the above fonnula, we can now apply the MINIMAX criterion defined in b. D. HURWICZ CRITERION This criterion represents a range of attitudes from the most optimistic to the most pessimistic. Under the most optimistic conditions, one would choose the action yielding: max max {v( Ai , (Jj } Ai { I } t/> Under the most pessimistic conditions, the chosen action cOlTesponds to: . max min {v( Ai , (Jj } A. A.. t { '1') } 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISIONTHEORY EDGARL. DE CASTRO PAGE5 " '.. ". ., : '..: ;.," .:j"...... . . . ..
  • 6. The Hurwicz criterionstrikes a balancebetween extremepessimism and extreme optimism by weighing the above conditions by respective I weights a and (1- a), where 0 < a <1. That is the action selected is t , that which yields:' . ., ma..'Xa m~ v( Ai , r/J ) + (1 - a )min v( Ai , r/J ) j j Ai { t/J j t/J j } .. [Note the above formulas represent the case where payoffs are expressed as profits] If a = 1, the decision rule is referred to as the MAXIMAX RULE, and if a = 0, the decision rule becomes the MAXIMIN RULE. For the case where the payoff represent costs, the decision rule is given by: min a min v( Ai , f/Jj ) + (1- a) max v( Ai , f/J ) j ~ { ~ ~ } E. BAYES' RULE Here, weasswne that the probabilities associated with each state of natureareknown.Let . . P{ljJ = ljJj} = Pj The action which minimizes (maximizes) the cost (profit) is selected. This is given by: The backward induction approach is used. With the aid of a decision tree, expected values are computed each time a round node is encountered and the above decision rule is utilized each time a square node is encountered ,i.e., a decision is made each time a square node is encountered. IIIIIIIIIIIII!IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISIONTHEORY EDGARL. DE CASTRO PAGE6 . . . .: .: . .~ . .,". ':". .,',,1, .. . ','. i.L ,.
  • 7. .. I .' t, F. EXPECTED VALUE-VARIANCE CRITERION This is an extension of the expected value criterion. Here we simultaneously maximi~e profit and minimize the variance of the profit. . . If Z represents profit as a random variable with variance q:, then the criterion is given by: maximize E(z) - Kvar(z) where K is any specified constant. If Z represents cost: .' . ./ minimize E(z) + Kvar(z) G. DECISION MAKING WITH EXPERIMENTATION In some situations, it may be viable to secure additional information tp revise the original estimates of the probability of occurrence of the state of nature. DEFINITION 1.2 : Preposterior Analysis considers the question of deciding whether or not it would be worthwhile to get additional information or to perfonn further experimentation. DEFINITION 1.3 : Posterior Analysis deals with the optimal choice and evaluation of an action subsequent to all experimentation and testing using the experimental results. /1/1/1111111111111111111111/111111111111111111111111111111111111111111111/111111/11111111/1111111111111111111/111111111111111111/1/11111/1111111111111111111111111111111111111111111111111/1/111111111/1/111111111111111111111/1/111111/11111111111 DECISION THEORY EDGAR L. DE CASTRO PAGE 7 " ".':.:::: >. " :.-:'.:.:..
  • 8. ---.......----- DEFINITION 1.4 : Prior probabilities are the initial probabilities assumed without the benefit of experiment~tion. Posterior probabilities refer to the revised probability values obtamed after experimentation. . - Let: Pj =prior probability estimate of event (Jj P{Zkl(Jj} = conditional probability of experimental (jutcome Zk P{(Jj IZk} = posterior probability of event (Jj The experimental results are assumed to be given by Zk, k = 1, 2, ... 1. The conditional probability can be considered to be a measure of the reliability of the ~xperiment. The idea is to calculate the posterior probabilities by combining the prior probabilities and the conditional probabilities of experimental outcome Zk. The posterior probabilities are given by: m L P{Zk l(Ji}P{(Ji} i=1 Once the posterior probabilities are calculated, the original problem can be viewed as a multiple stage/sequential DT problem. The first stage involves the decision of whether to perform additional experimentation or not. Once this is decided, the outcomes of the experiment are considered together with the original set of decision alternatives and events. 1111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111/1111111111111111111111111111111111111111111111111111111111111111 DECISIONTHEORY EDGARL. DE CASTRO PAGE8 . '. .
  • 9. DEFINITION 1.5 : A perfect infonnation source would provide, with 100% reliability, which of the states of nature wouldoccur. . >'. Define: EPPI = expected profit from a perfect information source EVPI = expected value of the perfect infonnation source EP = Bayes' expected profit without experimentation Then: ., EVPI = EPPI - EP where: 11 E VPI = L Pj * max { v( Ai , fjJ ) } j . j=1 Ai EVPI is easily seen as a measure of the maximum amount a decision maker should be willing to pay for additional infonnation. Define: EVSI = expected value of sample information ENOS = expected net gain from sampling CAI = cost of getting additional information 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISIONTHEORY EDGARL. DE CASTRO PAGE9 . '.' '. .. .... . .' .' ., . ..' ',' I"'. ," " .. ..
  • 10. , ., , Then: I , , " ,. - , ENGS = EVSI - CAI .. The information source would be viable if ENGS > O. II. CONTINUOUS DECISION THEORY. As previously mentioned, continuous decision theory problems refer to those where the number of alternatives and/or states of nature can be considered infmite. The optimization model in this case is given by: . max f(A) =J: v(A,t/J)htjJ(t/J)dt/J where: htjJ(t/J) = prior distribution function of the states of nature In the above model, it is assumed that no additional infonnation is available and the expectation is evaluated with respect to the prior distribution of the states of nature. If additional infonnation is available, we update the prior distribution of the states of nature by detennining its posterior distribution, which is nothing but the conditional distribution of the states of nature given the experimental outcome. Hence, the optimization converts to: IIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIHIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISIONTHEORY EDGARL. DE CASTRO PAGE 10 . ..... ..:...:.... ... . ;...:..' :.. .... . . . . ..
  • 11. ------- . max f(A) = f: v(A,f/J)ht/>IZ=z (f/J)df/J where: '.. .jI I hfj)IZ=z (rjJ) = conditional distribution of the state of nature' given the experimental outcome :! I I ..J hZIfj)(z) = conditional distribution of the experimental outcome given the state of nature hz (z) = marginal distribution function of the experimental outcomes where: LEIBNIZ' RULE LEIBNIZ' Rule is applied to find the derivative'of a function which contains integrals. Consider a function in one variable A: d b ig b db da - fa g(A,rjJ)drjJ=fa ~rjJ+g(A,b)--g(A,a)- ciA 8A ciA ciA 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 DECISION THEORY EDGAR L. DE CASTRO PAGE 11 '. '.: '. . :