Operations Research
Operations Research
GROUP – 2
ABHISHEK NIGAM
ANUPAM ROY
DISHA KOTECHA
MAHESH OGANIA
PRIYA PATIL
SANYA RAWAL
SHUBHRA SALGAONKAR
STATISTICAL DECISION THEORY
A general theory for the processing and use of statistical observations. In a broader
interpretation of the term, statistical decision theory is the theory of choosing an
optimal non-deterministic behavior in incompletely known situations.
Decision-making means a process which results in the selection from a set of
alternative courses of action, that course of action which is regarded to meet the
objectives of the decision problem more satisfactorily as compared to others.
Process of decision making includes: Defining the problem, Analysis of the problem,
Identification of alternatives, Evaluation of alternatives, Decision environment.
TERMS AND TERMINOLOGIES USED
Goals: Objectives which the decision maker wants to achieve by his actions.
Acts: Acts are the actions being considered by the agent or the decision maker.
Acts or alternatives- denoted by a1,a2,a3.
Events: Events are occurrences taking place outside the control of the agent.
Events- denoted by e1,e2,e3.
Pay off-measures the net benefit to decision maker, which results from a given combination.
Outcomes: Outcomes are the result of the occurrence of acts and events.
Opportunity loss or regret-the difference between pay off realized and the maximum
payoff which could have been realized if another strategy was chosen.
DECISION MAKING ENVIRONMENTS
This criterion is based on the principle of insufficient reason and was developed by Thomas
Bayels and supported by Simon de Laplace.
It is assumed that all states of nature will occur with equal probability
Probabilities of each state of nature is given by 1/( number of states of nature)
By using this criterion it is possible to convert a decision problem under certainty with equal
probability of each state of nature
ALGORITHM INVOLVED
Example:
A farmer wants to decide which of the three crops he should plant on his 100 Acre farm.
The profit from each is dependent on the rainfall during the growing seasons. The farmer has
categorized the amount of rainfall as high, medium, low. His estimated profit for each is show
in the table:
Rainfall Crop A Crop B Crop C
Assign equal probabilities i.e. 1/3. The expected pay off is calculated for each alterative:
E (Crop A)=1/3(8000)+1/3(4500)+1/3(2000)= 4833
E (Crop B)=4333
E (Crop C)=4666
Hence this criterion selects crop A.
MAXIMIN (MINIMAX) CRITERION
It suggests that the decision maker examines only the minimum payoffs of alternatives and
chooses the alternative whose outcome is the least bad.
DECISION MAKING
If v(ai, sj) is loss, then we select the action that corresponds to the following minimax criterion:
If v(ai, sj) is gain, then we select the action that corresponds to the following maximin criterion:
STEPS INVOLVED
1. Determine the objective (that is, whether it is a loss or gain we are interested in)
Row Min
40
81
82 Maximin
When α=0, the criterion is too pessimistic; when α=1, it is too optimistic. A value of α
between zero and one may be selected depending upon whether the decision-maker leans
towards pessimism or optimism. A value of α=1/2 seems to be a reasonable choice.
HURWICZ CRITERION EXAMPLE
Economy
Alternatives Growing Stable Declining
Bonds 40 45 5
Stocks 70 30 -13
Mutual Funds 53 45 -5
• The Regret criterion focuses on avoiding regrets that may result from making a
non‐optimal decision.
• Regret is defined as the opportunity loss r(a(i),θ(j)) to the decision maker if action
alternative a(i) is chosen and state of nature θ(j) happens to occur.
• Opportunity loss is the payoff difference between the best possible outcome under θ(j)
and the actual outcome resulting from choosing a(i)
• The best possible OL is zero (no regret) and the higher the OL value, the greater the
regret.
Steps to calculate
Assumptions
One more way of maximizing monetary value is to minimize the expected opportunity loss
or expected value of regret.
The expected opportunity loss (EOL) for a particular course of action is determined by
taking the difference between the payoff value of the most favourable course of action and
some other course of action.
Thus, opportunity loss can be obtained separately for each course of action by first obtaining
the best state of nature for the prescribed course of action and then taking the difference
between that best outcome and each outcome for those courses of action.
Pay Off Table
Profit in $000’s
Investment (States of Nature)
Choice Strong Stable Weak
(Action) Economy Economy Economy
Large factory 200 50 -120
Average factory 90 120 -30
Small factory 40 30 20
EVPI helps to determine the worth of an insider who possesses perfect information.
The expected value with perfect information is the amount of profit foregone due to
uncertain conditions affecting the selection of a course of action.
Given the perfect information, a decision-maker is supposed to know which
particular state of nature will be in effect.
HOW TO CALCULATE EVPI?
Payoffs
SOLUTION
First we calculate the Emv and find the maximum value which will be equal to EMwoPI
EVPI= 38-32 =6
THANK YOU