Decision trees

• A graphic representation of a problem which

facilitates making the right decision
 High 0,3
+ 1000
Production Mid 0,5
E +200
Low 0,2
Positive 0,6 -200
C

Test Discard
- 100 +50

Production
Negative 0,4 D -600
A
Discard
+50

Discard
+50
 High 0,3
+ 1000
Production EV = Mid 0,5
E +200
360
Low 0,2
Positive 0,6 -200
C

Test Discard
- 100 +50
EV =
B
236
EV =
136
Production
Negative 0,4 D -600
A
Discard
+50
EV =
50
Discard
+50

BACKWARD INDUCTION PROCEDURE

 Example
• A company is deciding whether to develop and launch a new product.
Research and development costs are expected to be $400,000 and there is
a 70% chance that the product launch will be successful, and a 30%
chance that it will fail. If it is successful, the levels of expected profits and
the probability of each occurring have been estimated as follows,
depending on whether the product’s popularity is high, medium or low:

• If it is a failure, there is a 0.6 probability that the research and

development work can be sold for $50,000 and a 0.4 probability that it will
be worth nothing at all.
 Games with nature
• randomly behaving competitor
• market
• stock market
• weather
• general conditions
 Games with nature – optimal
behaviour
Four decision paradigms:
• Wald paradigm
• Savage paradigm
• Hurwicz paradigm
• Laplace-Bayes paradigm
 Wald paradigm (maximin)
• the best decision is based on the worst possible
outcomes of every state

State A State B State C

Decision 1 4000 8000 6000

Decision 2 7000 3000 10000

 Savage paradigm(minimax)
• based on the matrix of alternative losses

State A State B State C

Decision 1 3000 0 4000

Decision 2 0 5000 0
 Hurwicz paradigm
• based on the pessimism coefficient λ є <0,1>
• h = λ * min + (1 – λ) * max
• if not specified otherwise we assume λ = 0,25

State A State B State C h

Decision 1 4000 8000 6000 7000
Decision 2 7000 3000 10000 8250
 Laplace-Bayes paradigm
• an equal probability of all the states is
assumed and expected value is calculated:
• EV = p * w

State A State B State C EV

Decision 1 4000 8000 6000 6000
Decision 2 7000 3000 10000 6666