Decision Theory and Decision Tree Analysis

You might also like

Download as ppt, pdf, or txt
Download as ppt, pdf, or txt
You are on page 1of 43

Course Code: FNCE 212

Course Name: Management Mathematics 2

Lecture 7: Decision Theory

P. Chimwani-Nengo
Devotional Meditation
Luke 19:1-10
•Then Jesus entered and passed
through Jericho. 2 Now behold, there was a man
named Zacchaeus who was a chief tax collector,
and he was rich. 3 And he sought to see who Jesus
was, but could not because of the crowd, for he
was of short stature. 4 So he ran ahead and
climbed up into a sycamore tree to see Him, for He
was going to pass that way.
Devotional Meditation
5
And when Jesus came to the place, He looked
up and saw him, and said to
him, “Zacchaeus, make haste and come down,
for today I must stay at your house.” 6 So
he made haste and came down, and received
Him joyfully. 7 But when they saw it, they
all complained, saying, “He has gone to be a
guest with a man who is a sinner.”
.
Devotional Meditation
8
Then Zacchaeus stood and said to the Lord,
“Look, Lord, I give half of my goods to the poor;
and if I have taken anything from anyone by false
accusation, I restore fourfold.” 9 And Jesus said
to him, “Today salvation has come to this house,
because he also is a son of Abraham; 10 for the
Son of Man has come to seek and to save that
which was lost.”
.
Points to Ponder
 Zacchaeus an important man in society but was
looking to know God. Are you seeking God? Your
status doesn’t matter …every human being needs to
have a relationship with God i.e. make peace with
God through His Son the Lord Jesus Christ 2 Cor 5:
17-21
2 Cor 5:17-21
Therefore, if anyone is in Christ, he is a new creation; old things have passed away; behold, all things have become new. Now all things are of God, who has reconciled us to Himself through Jesus Christ, and has given us the ministry of reconciliation, that is, that God was in Christ reconciling the world to Himself, not imputing their
trespasses to them, and has committed to us the word of reconciliation. Now then, we are ambassadors for Christ, as though God were pleading through us: we implore you on Christ’s behalf, be reconciled to God. For He made Him who knew no sin to be sin for us that we might become the righteousness of God in Him.
Points to Ponder
 Zacchaeus went through great lengths to do so risking his
reputation and societal standing. Don’t worry about what
people think or say, it’s your soul here and where you’ll spend
eternity and God holds you personally accountable for it.
 When Zacchaeus finally saw Jesus and had supper with Him,
Jesus changed/ transformed him- he became a new man,
heart, he became a true disciple, he truly repented and bore
the fruits of repentance i.e. restitution Jericho had a new Chief
of KRA  One who feared God- the society felt the difference
Points to Ponder
 You too can become a new man today and every day as you grow
in your relationship with Jesus.

 Jesus didn’t listen to public’s opinion of Zacchaeus …that he is a


sinner, a publican, a thief, corrupt etc….

 Remember Jesus mission and mandate is to seek and save the


lost irrespective of their state.
Email
pnengo@kabarak.ac.ke
Extending invitation for Bible Study Group
Further Consultation
1
1

Decision Theory: Definition


/
1
9
/
2 Decision making can be defined as a
3
process which results in the selection
from a set of alternative courses of
action, that course of action which is
considered to meet the objectives of
the decision problem more
satisfactorily than others as judged by
the decision maker.
1
1

Definition of Terms
/
1
9
/ Action space: is made up of all the
2
3 alternative courses of action from which
the final decision must be made.
The acts: these are the alternative courses
of action or strategies that are available
to the decision maker. The problem is to
choose the best of these alternatives in
order to achieve an objective. The
decision maker has control over the
choice of these.
1
1
/ Events or states of nature: the circumstances
1 that affect the outcome of the decision and are
9
/ beyond the control of the decision maker.
2
3
Outcomes: for each combination of act
(alternative) and event (state of nature) called
act-event combination there is an outcome.
This outcome may be expressed in terms of
profits/present values /some other non-
monetary measures. Outcomes are usually
referred to as pay-offs.
1
1
/ • Pay-off: each combination of a course of action
1
9 and an event is associated with a pay-off, which
/ measures the net benefit to the decision maker
2 that accrues from a given combination of
3 decision alternatives and events. Pay-offs are
also known as conditional profit values or
conditional economic consequences.

• Opportunity loss: the decision maker might


experience regret after the decision has been
taken and the events have occurred. Opportunity
loss is the difference between the pay-off of the
most favorable act (i.e. highest pay-off) and
some other act.
1
1
/
1
9
Types of Decision-making
/
2
3 1.Decision making under uncertainty: these refer to
situations where more than one outcome can result from
any single decision i.e. one decision several outcomes.

2. Decision making under certainty: whenever there exists


only one outcome for a decision we are dealing with this
category e.g. linear programming, transportation,
assignment and sequencing i.e. one decision one outcome.
1
1
/
1 3. Decision- making using prior data: it
9
/
occurs whenever it is possible to use
2 past experience (prior data) to develop
3 probabilities for the occurrence of each
event.
4. Decision-making without prior data:
no past experience exists that can be
used to derive outcome probabilities in
this case and the decision maker uses
his/her subjective estimates of the
probabilities of various outcomes.
1
1
/
1
9
Decision-making Under Uncertainty
/
2 The following are the decision criteria/rules for making decisions
3 under such situations:
1. Maximin criterion
2. Maximax criterion;
3. Minimax regret method
4. the Hurwicz method
5. the Laplace (equal likelihood) method
1
1

1. MAXIMIN Criterion
/
1
9
/
2 • This criterion is based on the “conservative approach”
3
which assumes that the worst possible is going to
happen.
• The decision maker considers each strategy and
locates the minimum pay off for each and then selects
that alternative which maximizes the minimum payoffs
i.e the best of the worst.
1
1

Example
/
1
9
/
2
3
The payoff table below shows the potential
profit and losses of three products A, B and C
in three market conditions: Boom, Steady
state and Recession state.
1
1

Pay-off Table Example


/
1
9
/ State of Nature
2
3
Alternative Boom *1 Steady *2 Recession *3

Product A +8 +1 -10

Product B -2 +6 +12

Product C +16 0 -26


1
1
/
1 • Applying the Maximin rule: Prepare a pay-off
9
/
matrix as follows.
2
3
*1 *2 *3
RowMinima
A 8 1 -10 -10
B -2 6 12 -2
C 16 0 -26 -26

For each strategy determine the minimum pay-


off (Row Minima).
Then select the maximum of the row minima.
Ranking by the MAXIMIN rule is B A C
1
1

MAXIMAX Criterion
/
1
9
/
2 This method is based on ‘extreme optimism’ the decision maker
3 selects that particular strategy which corresponds to the
maximum of the maximum pay off for each strategy i.e. the
best of the best.
Example 2
Rank the products A B and C applying the Maximax rule.
1
1
/
1
9 • Applying the Maximax rule: Prepare a
/ pay-off matrix as follows.
2
3 *1 *2 *3Row
Maxima Maximum
A 8 1 -10 8 of Row
B -2 6 12 12 Maxima
C 16 0 -26 16
For each strategy determine the
maximum pay-off (Row Maxima).
Then select the maximum of the row
maxima.
Ranking by the MAXIMAX rule is C B A
1
1

MINIMAX Regret Method


/
1
9
/
2 This method assumes that the decision maker will
3 experience ‘regret’ after he has made the decision and the
events have occurred. The decision maker selects the
alternative which minimizes the maximum possible regret.
Step 1: Prepare the regret table by considering the highest
pay-off for each column (state of nature) and subtracting
the others from this i.e. opportunity loss.
Step 2: Determine the row maxima and then select the
minimum of the row maxima i.e. minimize maximum
possible regret.
1
1
/
1
9 Applying the Minimax Regret method:
/ Prepare the regret matrix as follows.
2
3 *1 *2 *3Row Minimum
Maxima of Row
A 8 5 22 22 Maxima
B 18 0 0 18
C 0 6 38 38
For each strategy determine the maximum
regret and then select the minimum of the
maximum possible regret.
Ranking by the MINIMAX Regret B A C
1
1

HURWICZ Method
/
1
9
/
2 This method uses the concept of coefficient of
3
optimism, α (or pessimism, 1- α) introduced by L.
Hurwicz.
The decision maker takes into account both the
maximum and minimum pay-offs for each
alternative and assigns them weights according to
his degree of optimism (or pessimism).
1
1
/
1 The alternative which maximizes the sum of
9
/ these weighted payoffs is then selected
2
3
Weighted pay-off = (maximum pay-off) * α +
(minimum pay-off) *(1-
α)
Example 4:
Rank the products A, B and C by applying the
Hurwicz method. Let coefficient of optimism
α = 0.7.
1
1

Solution
/
1
9
/
2
3
Product A: (8 x 0.7) + ( -10 x 0.3) = 2.6
Product B: (12 x 0.7) + ( -2 x 0.3) = 7.8
Product C: (16 x 0.7) + ( -26 x 0.3) = 3.4

• Ranking using Hurwicz method; B C A


1
1

Laplace or Equally Likely Method


/
1
9
/
2 • This method uses all the information by assigning
3
equal probabilities to the possible payoffs for each
action and then selecting that alternative which
corresponds to the maximum weighted expected pay
off.
If there are n possible pay-offs then the probability
assigned is 1/ n
1
1

Example 5: Ranking By Laplace


/
1
9
/
2 Because there are 3 possible payoffs p= 1/3
3
Product A: (8 x 1/3) + (1 x 1/3) + (-10 x 1/3) = -0.33
Product B: (-2 x 1/3 ) + (6 x 1/3) + (12 x 1/3) = 5.33
Product C: (16 x 1/3) + (0 x 1/3) + (-26 x 1/3) = -3.33
• Ranking using Laplace method; B A C
1
1
/
1
Exercise
9 Pay-offs on Events ‘000s
/
2
3
Alternative A B C D

X 8 0 -10 6

Y -4 12 18 -2

Z 14 6 0 8
1
1
/
1 Determine which alternative the
9
/ businessman would choose if he adopts:
2
3 a) the MAXIMIN criterion
b) the MAXIMAX criterion
c) the MINIMAX REGRET method
d) the HURWICZ method, α = 0.7
e) the LAPLACE method
a) MAXIMIN criterion Alternative Z

Answers
b) by MAXIMAX criterion Alternative Y

c) by MINIMAX REGRET method Alternative Y or Z

d) by HURWICZ method, α = 0.7 Alternative Y

e) by LAPLACE method Alternative Z


DECISION TREES AND SUB
SEQUENTIAL DECISIONS

11/19/23
1
1
/
1
9
Introduction
/ • A decision tree is a graphic display of various
2
3 decision alternatives and the sequence of events as
if they were branches of a tree.
The symbol and indicates the decision
point and the situation of uncertainty or event
respectively.
• The node depicted by a square is a decision node
where choices exist between alternatives and
managerial decisions are made based on estimates
and calculations of the returns expected.
1
1
/ • Branches out of a decision point represent
1
9 the immediate mutually exclusive alternative
/
2 acts/options open to the decision maker.
3
• The outcome nodes are depicted by a circle
where the events depend on probabilities.
branches out of an event point represent all
possible situations/events. These events are not
fully under the control of the decision maker.
When probabilities of various events are known,
they are written along the corresponding
branches.
1
1
/
1 Procedures
9 • Forward pass: Show all decision points and outcome
/
2 done from left to right, concentrate first on the logic of
3 the problem and on probabilities or values involved.
• Backward pass: This stage is worked from right to
left. Analysis of the expected values at various points
is done so the correct decision i.e. with the highest
expected value is made.
• The expected value for a decision is the highest pay
off value whereas the E.V for an outcome is the
summation of probability multiplied by the pay off
value of each branch.
1
1
/ The Expected Monetary Value (EMV)
1
The expected pay off Criterion
9
/ • (profit) associated with a given
2
3
combination of actions and events is obtained by:
Expected Pay-off = (Pay-off act-event combination) *
(probability of occurrence of event)
• Expected Monetary Value of an act is the sum of all
expected conditional profits associated with that act.
i.e. EMV = ∑ expected (conditional) pay-offs
• The decision maker chooses the act which corresponds
to the highest EMV. In so doing incorporates risk and
uncertainty to his decision making.
1
1

Example
/
1
9
/ A manager has a choice between
2
3 1. A risky contract promising KShs 7 million with
probability 0.6 and KShs 4 million with probability 0.4
and
2.A diversified portfolio consisting of two contracts with
independent outcomes each promising KShs 3.5 million
with probability 0.6 and KShs 2 million with probability
0.4
• Can you arrive at the decision using EMV method? This
is best done using decision tree analysis.
1
1
/
1 Example
9
/ • Kauzi Agro Mills (KAM) Ltd is considering whether to
2 enter a very competitive market. In case KAM decided
3 to enter this market it must either install a new forging
process or pay overtime wages to the entire workers.
• In either case, the market entry could result in high
sales; medium sales; low sales and no sales.
• Suppose the management of KAM has estimated that if
they enter the market there is a 60% chance of their
stakeholders approving the installation of the new forge.
(This means that there is a 40% chance of using
overtime).
1
1
/
1 Kauzi Agro Mills Ltd
9
/ • A random sample of the current market structure
2 reveals that KAM has a 40% chance of achieving high
3 sales; a 30% chance of achieving medium sales; a 20%
chance of achieving low sales and a 10% chance of
achieving no sales.
• Further, market analysts of KAM have indicated that a
high level of sales will yield KShs 1,000,000 profit; a
medium level of sales will result in a KShs. 600,000
profit a low level of sales will result in a KShs 200,000
profit and a no sales level will cause KAM a loss of
KShs 50,000 apart from the cost of any equipment.
1
1

KAM Ltd ctd…


/
1
9
/
2 • Entering the market will require a cash outlay of
3
either KShs 300,000 to purchase and install a
forge or KShs 100,000 for overtime expenses
should the second option be selected.
Required: Draw the appropriate decision tree
diagram and use it to determine what decision
KAM management should make regarding
whether or not to enter the market.
1
1
/
1 Make Own Notes
9
/ Advantages of Decision Trees
2
3

Limitations of Decision Trees

You might also like