100% found this document useful (5 votes)
279 views

Basic Decision Tree

Here are the steps to solve this problem: P(AC) = 0.7 P(CD) = 0.4 P(AC and CD) = 0.2 To find P(CD | AC), we use the definition: P(CD | AC) = P(AC and CD) / P(AC) Plugging in the values: P(CD | AC) = 0.2 / 0.7 = 0.286 Therefore, the probability that a car has a CD player given that it has AC is 0.286 or 28.6%

Uploaded by

ratu nida
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
100% found this document useful (5 votes)
279 views

Basic Decision Tree

Here are the steps to solve this problem: P(AC) = 0.7 P(CD) = 0.4 P(AC and CD) = 0.2 To find P(CD | AC), we use the definition: P(CD | AC) = P(AC and CD) / P(AC) Plugging in the values: P(CD | AC) = 0.2 / 0.7 = 0.286 Therefore, the probability that a car has a CD player given that it has AC is 0.286 or 28.6%

Uploaded by

ratu nida
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 48

Decision Tree

Choice of the Best Alternative

Manahan Siallagan

Problems of Choice

C1

O1

C2

O2

e
b
e
Th
2

ay
w
st

D : a decision maker
C : possible courses of action

(alternatives)
O1 : desirable outcome;
O2 : undesirable outcome

Complex Rational Choice


The information and insight
needed to predict the
consequences of each option
Incomplete information
The other side may influences the

consequences (Strategic/Interactive
Decision).

It is assumed that we can calculate


probability of the consequences
3

I cant see

Decision Analysis Problems


Decision analysis is designed to address

decision making in the face of great


uncertainty
Introducing new product into
marketplace:

What will the reaction of potential

customers ? Competitors?

Investing in securities :
How is the economy?
How about interest rates?

Selecting the mix of crops and livestock

for the upcoming season :

What will be the weather conditions?

Drilling for oil in particular location :


How likely is there to be oil in that

location?
How much?

Tools: Decision tree

Why Decision Tree?


It can help a decision maker to develop a

clear view of the structure of a problem


anda make it easier to determine the
possible scenarios that can result if a
particular course of action is chosen.
Decision trees can also help a decision
maker to judge the nature of the
information that needs to be gathered in
order to tacke a problem.
It can be an excellent medium for
communicating one persons perception of
a problem to other individuals.

Model of Decision Tree


C1
D
C2

E1,1

p1,1

O1

E1,2
E2,1

p1,2 O
2
p2,1 O1

E2,2

p2,2

O2

D : a decision maker
C : possible courses of action (alternatives)
O1 and O2 : possible outcomes/consequences/payoffs
Ei,j : Events (State of Natures/SON)
6

pi,j : probabilities

Structure of Decision Tree


Decision Node
Alternatives available for decision maker to

choose;
Situation controllable by decision maker.

Alternatives of actions
7

Structure of Decision Tree


Event Node
Events may happen after every action made

by decision maker;
Uncontrollable by decision maker;
Decision maker only has information about
probability of each event no complete
information.
E1
E2

p1
p2

An action
E3
8

p3
Events

Building Decision
Tree
1. Identify what decisions should be made

by DM;

What are the

first

decision, and
decisions to be made?

next

2. Identify what SON happen after each

decision;
3. Draw decision node and event node
(SON);
4. Complete information about probabilities;
5. Complete information about payof.
9

Goferbroke Company Case

Goferbroke
Company(1)
Max Flyer is the founder of and sole owner of the

Goferbroke Company, which develops oil wells in unproven


territory. Maxs friends refer to him afectionately as a
wildcatter. However he prefers to think himself as an
entrepreneur. He has poured his life savings into the
company in the hope of making it big with a large strike of
oil.
Now his chance possibly has come. His company has
purchased various tracts of land that larger oil companies
have spurned as unpromising even though they are near
some large oil fields. Now Max has received an exciting
report about one of these tracts. A consulting geologist
has just informed Max that he believes there is one
chance in four of oil there.
Max has learned from bitter experience to be skeptical
about the chances of oil reported by consulting geologist.
Drilling for oil on this tract would require an investment of
about $100,000. If the lands turns out to be dry (no oil),
the entire investment would be lost. Since his
company doesnt have much capital left, this lost would be

Goferbroke Company(2)
On the other hand, if the tract does contain oil, the

consulting geologist estimates that there would be


enough there to generate a net revenue of
approximately $800,000, leaving an approximate profit
of:
Profit if find oil = Revenue if find oil Drilling cost
= $800,000 - $100,000
= $700,000

There is another option that another oil company has

gotten wind of consulting geologists report and so has


ofered to purchase the tract of land from Max for
$90,000. This is very tempting. This too would provide a
welcome infusion of capital into the company, but
without incurring the large risk of a very substansial loss
of $100,000.

The Goferbroke Company Problem


Decision that must be taken:

Should Max sell his land or doing drilling?


Alternative:
1. Sell land
2. Drilling
Possibility event that could happen (state of nature):
Found Oil
No Oil (dry)
Payoff Table (Information from the case)
Payoffs

If you are Max, which alternatives that you would


choose?

The Maximax Decision


Criterion
Focus only on the best that can happen the

maximax criterion always chooses the decision


alternative that can give the largest possible
payoff Total Optimist
Identify the maximum payoff from any SoN for
each decision alternative
Find the maximum of these maximum payoffs
and choose the corresponding decision
alternative
Weakness : abandoning prior probability and
abandoning other payoff beside only the biggest.
Alternative
Drill for oil
Sell the land

State of Nature
Oil
Dry
$700
-$100
$90
$90

Max in Row
700
90

Maximax

The Maximin Decision


Criterion
Focus only on the worst that can happen to us

(total pessimist)
Identify the minimum payoff from any SoN for each
decision alternatives
Find the maximum of these minimum payoffs and
choose the corresponding decision alternative
Weakness : abandoning prior probability and
abandoning other payoff beside only the maximin.

Alternative
Drill for oil
Sell the land

State of Nature
Oil
Dry
$700
-$100
$90
$90

Min in Row
-$100
90

Maximin

The Maximum Likelyhood


Criterion
Focus on the most likely state of nature.
Identify the state of nature with the largest

prior probability;
Choose the decision alternative that has
the largest payoff for this state of nature.
Weakness: abandoning payoff, that
actually payoff maybe very big.
Alternative
Drill for oil
Sell the land
Prior Probability

State of Nature (thousands)


Oil
Dry
$700
-$100
$90
$90
0.25
0.75

Step 1 : Maximum

Step 2 : Maximum

Bayes Decision Rule


Bayes Decision Rule choose best alternative by considering

entire information that being owned by doing steps mentioned:


Calculate Expected Value for every decision alternative
EV = (prior prob x payoff)
Choose the decision alternative that has the largest Expected
Value

Advantage :
Considering entire information (alternatives, payoffs, and prior
probabilities);
In long term, if the decision occur sequential, then this criteria
will resulting payoff that mostly probably happen.

Alternative
Drill for oil
Sell the land
Prior Probability

State of Nature (thousands)


Oil
Dry
EV
$700
-$100
$90
$90
0,25
0,75

100
90

Maximum

Decision Trees of Goferbroke


Decision tree is decision making help

tools that could describe entire


alternatives with whole events that may
happen (SoN).
Showing : Alternatives, SoN, Prior
Probability, and Payoff.
Using Bayes Decision Rule to choose the
best action.

DT of Goferbrokes Case
Decision:
Drill or Sell the Land

SON
Oil or Dry

Decision nodes

Drill
-100
Sell
90

DT of Goferbrokes Case
Event nodes
0.25

Oil

800

Drill
-100
Sell
90

Dry

0,75
0

DT of Goferbrokes Case
Payof
0.25

Oil
Drill
-100
Sell
90

800
Dry

0,75
0

700

-100
90

DT of Goferbrokes Case
Payof
0.25

Oil

800

Drill
-100
100

100

0,75
0

Sell
90

Dry

90

700

-100
90

Expected payoff
= MAX [100,90]
= 100

Expected Value (EV) per event node;

Action: Drill

100=(0.25*700) + (0.75*(-100))

Decision Analysis:
New Information or Posterior
Probability

23

Process in Revising Decision tree


Prior probability
New information

Posterior
probability

24

Assessing Probability
There are three approaches to assessing the

probability of an uncertain event:


1. a priori classical probability
probability of occurrence

X
number of ways the event can occur

T
total number of elementary outcomes

2. empirical classical probability


probability of occurrence

number of favorable outcomes observed


total number of outcomes observed

3. subjective probability
an individual judgment or opinion about the probability of
occurrence

Computing Joint and


Marginal Probabilities
The probability of a joint event, A and B:

number of outcomes satisfying A and B


P( A and B)
total number of elementary outcomes
Computing a marginal (or simple) probability:

P(A) P(A and B1 ) P(A and B 2 ) P(A and Bk )


Where B1, B2, , Bk are k mutually exclusive and

collectively exhaustive events

Joint Probability Example


P(Red and Ace)

number of cards that are red and ace 2

total number of cards


52

Type

Color
Red

Black

Total

Non-Ace

24

24

48

Total

26

26

52

Ace

Marginal Probability Example


P(Ace)

P( Ace and Re d) P( Ace and Black )

Type

Color

2
2
4

52 52 52

Red

Black

Total

Non-Ace

24

24

48

Total

26

26

52

Ace

Computing Conditional Probabilities


A conditional probability is the probability of

one event, given that another event has


occurred:

P(A and B)
P(A | B)
P(B)

P(A and B)
P(B | A)
P(A)

The conditional
probability of A
given that B has
occurred
The conditional
probability of B
given that A has
occurred

Where P(A and B) = joint probability of A and B


P(A) = marginal probability of A
P(B) = marginal probability of B

Conditional Probability
Example

Of the cars on a used car lot, 70% have


air conditioning (AC) and 40% have a CD
player (CD). 20% of the cars have both.

What is the probability that a car has a CD

player, given that it has AC ?

i.e., we want to find P(CD | AC)

Conditional Probability
Example

(continued
Of the cars on a used car lot, 70% have air conditioning
) (AC) and
40% have a CD player (CD).
20% of the cars have both.
CD

No CD

Total

AC

0.2

0.5

0.7

No AC

0.2

0.1

0.3

Total

0.4

0.6

1.0

P(CD and AC) 0.2


P(CD | AC)

0.2857
P(AC)
0.7

Conditional Probability
Example

(continued
)
Given AC, we only consider the top row (70% of the cars). Of
these, 20% have a CD player. 20% of 70% is about 28.57%.
CD

No CD

Total

AC

0.2

0.5

0.7

No AC

0.2

0.1

0.3

Total

0.4

0.6

1.0

P(CD and AC) 0.2


P(CD | AC)

0.2857
P(AC)
0.7

Using Decision Trees


.2
.7
D
C
Has

Given AC or no
AC:

Ha

All
Car
s

)=
C
A
P(
C 0.7
A
s

Do
e
not s
hav
AC e

P(A
0.3 C)=

Do e
s
not
hav
e
CD

.5
.7

.2
.3
D
C
Has
Do e
s
not
hav
e
CD

.1
.3

P(AC and CD) =


0.2
P(AC and CD) =
0.5

P(AC and CD) =


0.2
P(AC and CD) =
0.1

Using Decision Trees


.2
.4
C
A
Has

Given CD or no
CD:

)=
D
P(C
D 0.4
C
s
Ha

All
Car
s

Do
e
not s
hav
CD e

P(C
0.6 D)=

Do e
s
not
hav
e
AC

.2
.4

.5
.6
C
A
Has
Do e
s
not
hav
e
AC

.1
.6

(continued
)
P(CD and AC) =
0.2
P(CD and AC) =
0.2

P(CD and AC) =


0.5
P(CD and AC) =
0.1

Bayes Theorem
Bayes theorem is used to revise previously calculated
probabilities after new information is obtained

P(A | B i )P(B i )
P(B i | A)
P(A | B 1 )P(B 1 ) P(A | B 2 )P(B 2 ) P(A | B k )P(B k )
where:

Bi = ith event of k mutually exclusive and


collectively
exhaustive events
A = new event that might impact P(Bi)

Goferbrokes Case
Continued
Survey by geologist will provide more

accurate information about P(oil);


How if Max has to decide two alternatives:
1.
2.

Do survey before drill/sell


Drill/sell without Survey

Events:
Do Survey

FSS : Favorable Seismic Sounding : Oil is fairly


likely
USS : Unfavorable seismic sounding: Oil is quite
unlikely.

Drill or Sell

Oil
Dry

30000

Max`s
Experience
P(state) prior; which is P(Oil)=0.25 & P(Dry)=0.75;
P (finding|state) being known based on Maxs
experiences; which is
P(FSS|Oil)=0.6,
P(USS|Oil)=0.4,
P(FSS|Dry)=0.2, and
P(USS|Dry)=0.8
Which:

State : Oil and Dry;


Finding : FSS and USS;
FSS : favorable seismic sounding; oil is fairly likely;
USS : unfavorable seismic sounding; oil is quite unlikely.

P(FSS Oil) = 0,6


P(USS Oil) = 0,4
P(FSS Dry) = 0,2
P(USS Dry) = 0,8

Posterior Probability Formula


P(FSS|Oil) = P(FSS&Oil) / P(Oil)
P(FSS|Dry) = P(FSS&Dry) / P(Dry)
P(Oil|FSS) = P(Oil&FSS) / P(FSS)
P(Dry|FSS) = P(Dry&FSS) / P(FSS)
P(Oil|USS) = P(Oil&USS) / P(USS)
P(Dry|USS) = P(Dry&USS) / P(USS)
P(FSS&Oil) = P(Oil&FSS) Law of Probability

Contingency
table
FSS

USS

Oil

0.15

0.1

0.25

Dry

0.15

0.6

0.75

0.3

0.7

P(FSS|Oil) = P(FSS& Oil) / P(Oil)


Since :
P (Oil) = 0.25; P(FSS|Oil) = 0.6
Then
P(FSS & Oil) = P(FSS|Oil) x P(Oil)
= 0.6 X 0.25
= 0.15
Do same step for find P (FSS&Dry) = 0.15
So :

P(FSS) = P(FSS&Oil)+P(FSS&Dry)
= 0.3

Posterior Probability Formula


(contd)
P(Oil|FSS) = P(Oil&FSS) / P(FSS)
Since : P(FSS) = 0.3; from contigency table
P(Oil&FSS) = P(FSS&Oil) = 0.15
Then : P(Oil|FSS) = 0.15 / 0.3 = 0.5
Do same step for P(Oil|FSS); P(Dry|FSS); P(Oil|USS);
P(Dry|USS)

FSS

USS

Oil

0.15

0.1

0.25

Dry

0.15

0.6

0.75 P(Dry|FSS)= P(Dry&FSS) = 0.15 = 0.5

0.3

0.7

P(Oil|FSS)= P(Oil&FSS) = 0.15 = 0.5


P(FSS)
0.3

P(FSS)

Do same step for P(Oil|USS); P(Dry|USS)

43

0.3

Leveled Decision Analysis


Decision T ree for Goferbroke Co. Problem (With Survey)
0,143
Oil

P(USS)

670
Drill

800

-100

-15,714

0,7
Unfavorable

0,857
Dry

-30

670

P(Dry|USS)
-130

Expected payoff
= MAX [123,100]
Do S urvey
= 123

P(Oil|USS)

-130

60
Sell
60
90

60
0,5

123

P(FSS)

670
Drill

800

-100

270

0,3
Favorable

0,5
Dry

670

P(Dry|FSS)
-130

1
0

P(Oil|FSS)

Oil

-130

270

Sell

123

60

Drill
-100

100

90

60

0,25
Oil

P(Oil)

800

700

0,75
Dry

P(Dry)

-100

No S urvey
1
0

700

-100

100
Sell
90
90

90

Posterior Probability
Given:
P(state)=prior probability: P(oil) and P(dry)
P(finding|state) = Maxs experience on probabilities of finding (FSS or USS)

could occur if some SoN (oil or dry) has been already happened.

P(FSS|oil)

0.25*0.6=0.15
Oil and FSS

P(Oil and FSS)


P(oil)

0.15/0.3=0.5
Oil, diket FSS

P(oil|FSS)

P(state|finding)

Prob. posterior

P(Oil and USS)


0.25*0.4=0.1
Oil and USS

P(USS|oil)

0.1/0.7=0.14
Oil, diket USS

P(oil|USS)

P(finding and state)


P(FSS|dry)

0.75*0.2=0.15
Dry and FSS

P(Dry and FSS)

P(dry)

0.15/0.3=0.5
Dry, diket FSS

P(dry|FSS)

P(Dry and USS)


P(USS|dry)

0.75*0.8=0.6
Dry and USS

0.6/0.7=0.86
Dry, diket USS

P(dry|USS)

Thank You

47

Exercise
Von Holt was a general marketing manager. He was assigned

to assess the prospect of a new product to be manufactured.


The approximately production cost was USD 90,000. If it had
been sold out then it would gain revenue USD 500,000. In
other side, if he didnt decide to produce then the company
would loss nothing. Based on his experience, he knew that 0.2
was the probability that the market would accept the product.
Even though, he wasnt sure whether it represented the
current data. He had another option to run market research
and collect the last information, but it took investments as
many as USD 10,000. Again, based on experience, he had
known that if the customer accepted the product then the
survey would exhibit that the result was favorable with
probability 0.6. However, the probability would reduce to 0.1,
if the customer rejected.
Now, help Holt to decide the best strategy! . and dont
forget to :
Draw the tree diagram.
Show us how did you get the probability

You might also like