2 Modeling Tools For Decision Analysis
2 Modeling Tools For Decision Analysis
Analysis
2.1 Decision Table (payoff table)
2.2 Decision Tree
2.3 Influence Diagrams
Outcomes
States of Nature
Alternatives
Decision Problem
1
Elements of a Decision
How do we structure the problem?
(1) Objectives and means
(2) Alternatives to choose between
(3) Uncertainty in Events and Outcomes
(4) Consequences of the Decision
Outcomes
States of Nature
Alternatives
Decision Problem
2.1 Decision Table
(or pay-off table)
It is possible to represent the type of problem
we have been examining in a table called
decision matrix or pay-off table.
Alternatives
States of Nature Alternative 1 Alternative 2
State 1 Outcome 1 Outcome 3
State 2 Outcome 2 Outcome 4
(3/4)
States of Nature
Alternatives State 1 State 2
Alternative 1 Outcome 1 Outcome 2
Alternative 2 Outcome 3 Outcome 4
transposed
Alternatives
States of Nature Alternative 1 Alternative 2
State 1 Outcome 1 Outcome 3
State 2 Outcome 2 Outcome 4
1) The Elements of a decision
table (4/4)
2) Example 1: TOM BROWN
INVESTMENT DECISION (1/5)
Tom Brown has inherited $1000.
He has to decide how to invest the money for one
year. The return on each investment depends on
the (uncertain) market behavior during the year.
A broker has suggested five potential investments.
Gold
Junk Bond
Growth Stock
Certificate of Deposit
Stock Option Hedge
2) Example1 : TOM BROWN
Solution (2/5)
Construct a payoff table.
Select a decision making criterion, and
apply it to the payoff table.
Identify the optimal decision.
Evaluate the solution.
S1 S2 S3 S4 Criterion
D1 p11 p12 p13 p14 P1
D2 p21 p22 p23 P24 P2
D3 p31 p32 p33 p34 P3
2) Example 1: The Payoff Table
(3/5)
DJA is up more DJA is up DJA moves DJA is down DJA is down more
than1000 points [+300,+1000] within [-300, -800] than 800 points
[-300,+300]
Decision
Decision
Alternatives:
1. status quo: no change
2. extend: extending their production line buying a new machine
3. build: building a new production hall with new equipment
4. cooperate: finding additional business parters for production
Uncertainty involved:
Market reaction: after the decision, the sales can increase or
decrease.
Consequences:
Expected profit, shown in decision table on the next slide
In-class Example (2/2)
2.2 Decision tree
It is frequently useful to represent the type of problem we
have been examining in graphical form by constructing
what is known as a decision tree.
The tree diagram shows the logical progression that
occurs over time in terms of decisions and
outcomes and is particularly useful in sequential decision
problems - where a series of decisions need to be made
with each, in part, depending on earlier decisions and
outcomes.
Decision tree represents the decision problem in terms of chains of
consecutive decisions and chance events.
Time proceeds from left to right.
Uncertainties associated with chance events are modeled by
probabilities.
1) Components of Decision Tree
(1/2)
1) Characteristics of a decision
tree (2/2)
State 1
Outcome 1
1
State 2
Outcome 2
State 1
Outcome 3
2
State 2
Outcome 4
Decision Node
States of Nature
Small Box Medium Box Large Box
Decisions Office Office Office
Sign with Movie
$200,000 $1,000,000 $3,000,000
Company
Sign with TV
$900,000 $900,000 $900,000
Network
Prior
0.3 0.6 0.1
Probabilities
Using Expected Return Criteria
(4/8)
EVmovie=0.3(200,000)+0.6(1,000,000)+0.1(3,000,000)
= $960,000
EVtv =0.3(900,000)+0.6(900,000)+0.1(900,000)
= $900,000
Therefore, using this criteria, Jenny should select the
movie contract.
Solved Payoff Table
(5/8)
States of Nature
Decisions
Small Box Medium Large Box
EMV
Office Box Office Office
Sign with Movie
$200,000 $1,000,000 $3,000,000 $960,000
Company
Sign with TV
Network
$900,000 $900,000 $900,000 $900,000
Prior
0.3 0.6 0.1
Probabilities
Jenny Lind - Decision Tree
(6/8)
Small Box Office
$200,000
RULE 3: NO FORGETTING.
Any information known when a decision is made must
be remembered perfectly when any subsequent decision
is made. This implies that between every pair of
decisions in an influence diagram, there must be a
directed path that includes only decision nodes. And, if
there is an arrow from an uncertainty to a decision,
there must also be an arrow from that uncertainty to
any subsequent decision.
3) Developing Influence
Diagrams: an Example (1/4)
The Quality Wilton Ltd. produces carpets. A special type of carpet they
produce is sold for a price of 32 per square meter in the U.K. The
product of the company is considered to fall into the category of luxury
commerce item in the United Arab Emirates (U.A.E.) and sold for a price
of 56.
Since the product is accepted to be a luxury item, the prices are
influenced by the state of economy in export country. The company
management is expecting a recession (decline) in the U.A.E. economy. It
is stated by company authorities that the next 12 months are important
for the orders placed. There is an economical recession (decline)
anticipation based on local inflation and low commercial values. The
company will export its product immediately or it will wait for
another 6 or 12 month to do it.
The executives of the carpet company foresee three different scenarios for
the U.A.E. These are stable economical conditions, slightly worse
condition and severely bad conditions. The pay-off matrix for the
problem and probabilities assigned by the company authorities for
economical state are given in Table 2.
3) Developing Influence
Diagrams: an Example (2/4)
Table 2. Pay-off matrix for Quality Wilton Ltd. ()
(3/4)
Influence Diagram
for Quality Wilton Ltd. (4/4)
4) Gradual Development of
Influence Diagrams
Two basic strategies:
Start with outcomes and model towards
decisions and events;
Gradually add more and more detail.
4) Example 1: Gradual
Development (1/3)
4) Example 1: Gradual
Development (2/3)
4) Example 1: Gradual
Development (3/3)
5) Example 2: Multiple Objectives
(1/1)
Two
Objectives
Single
Objective
6) Example 3: Intermediate
Calculations (1/1)
7) In-class Exercise 3: Tractor
Buying (1/2)
Your uncle is going to buy a tractor. He has two
alternatives:
1. A new tractor (17 000 )
2. An used tractor (14 000 )
The engine of the old tractor may be defect, which is
hard to ascertain. Your uncle estimates a 15 %
probability for the defect.
If the engine is defect, he has to buy a new tractor
and gets 2000 for the old one.
Before buying, your uncle can take the old tractor to
a garage for an evaluation, which costs 1 500 .
If the engine is OK, the garage can confirm it without
exception.
If the engine is defect, there is a 20 % chance that the
garage does not notice it.
7) In-class Exercise 3: Tractor
Buying (2/2)
States of Nature
Good Economic Conditions Poor Economic Conditions
Decisions (Probability=0.6) (Probability=0.4)
Apartment
$100,000 $60,000
building
Office
$200,000 $-80,000
building
Warehouse $60,00 $20,000
The end