Professional Documents
Culture Documents
By
Okite Moses
What is a Decision Tree?
• It is a Visual Representation of Choices, Consequences,
Probabilities, and Opportunities.
• It’s a Way of Breaking Down Complicated Situations
Down to Easier-to-Understand Scenarios.
• Can be used instead of a table to show alternatives,
outcomes, payoffs and calculate Expected Monetary
Value
Therefore a Decision Tree is…
A diagrammatic/graphical tool for describing;
5
Decision Trees analysis Steps
Calculate the tree values working from
the right side back to the left,
Calculate the values of uncertain
outcome nodes by multiplying the value
of the outcomes by their probability (i.e.,
expected.
Determine the best decision for the tree
by starting at its root and going forward
6
Example Decision Tree
Chance
Event 1
node
Decision ision1 Event 2
De c
node Event 3
Deci
sion
2
Note:
1. Create the decision tree from left to right
2. Solve the decision tree from right to left. This involves
calculating the expected value at each step
A larger tree diagram
DecisionAnalysis 8
Illustration of a Decision Tree
1. A decision point is shown by a rectangle
2. Alternatives available at a decision point are
shown as decision branches (DB).
DB 3. At the end of each DB, there
can be two or more chance
CB events shown by a node
20% chance branches (CB).
55%
Decision Chance events must be mutually
point 25%
exclusive and exhaustive (total
probability = 1).
20% + 55% + 25% = 100% = 1 as a
ratio
4. At the end of each branch is an endpoint shown as a triangle
where a payoff will be identified. 9
Example 1
• A bus operator has a choice of entering into a maintenances
contract for a new bus he has acquired that will cost him
$2,500. With the contract in operation, there is a 0.1
probability of a break down over the contract period but
without the contract, the probability is 0.3. The operator
reckons that if he can avoid a breakdown, his profits will be
$20,000 but this will drop to $10,000 in the event of such a
breakdown.
Required:
a) Construct a Decision Tree for this scenario
b) Should he enter into the contract?
So
EMV = (1,000 + 18,000) – lu
Break down ti
2,500 on
Profit = 10,000 x 0.1 = $1000
= $16,500
0.1
No Break down
0.9 Profit = 20,000 x 0.9 = $18,000
a ct
tr
o n 500
C ,
$2 Break down
No Profit = 10,000 x 0.3 = $3,000
co 0.3
$0
nt
ra
ct
0.7
No Break down
Profit = 20,000 x 0.7 = $14,000
No plant 0 0 0
• TV Network Payout
– Flat rate - $900,000
• Probabilities
– P(Small Box Office) = 0.3
– P(Medium Box Office) = 0.6
– P(Large Box Office) = 0.1
Solution – Decision Tree
Small Box Office
$200,000
0.1
ER Large Box Office
$3,000,000
960,000
0.1
Large Box Office $900,000
Buil
d, $ Successful deployment
a de 5 00
,000
pgr Impact = $0
U 0.6
St 0.4 Unsuccessful deployment
ay
le Impact = $2million
ga
cy
$100,
000
Obviously, if demand exceeds 500, you will sell all 500. On the
other hand, if demand is under 500, you will have leftover units.
These leftover items can disposed off for $7 each ($3 loss, the
dealer will no longer buy these leftover units from you).
What’s your decision?
26
Example 5...
Suppose you have 500 units of X in
Demand: X 300 400 500 600 stock, purchased for $10 each.
Pr(X) 0.30 0.45 0.20 0.05 Dealer sales price: $14, self sale
price: $16 with salvage value: $7.
Start with the tree having 2 branches
Dealer
Sale (DB) at the decision point.
3 0%
00, 45% There are no chance events in the
3 400,
Self sale
500, 20% dealer sale branch,
600
, 5% For the self sale, there are 4 mutually
exclusive possibilities.
Example 5...
Suppose you have 500 units of X in
Demand: X 300 400 500 600 stock, purchased for $10 each.
Pr(X) 0.30 0.45 0.20 0.05 Dealer sales price: $14, self sale
price: $16 with salvage value: $7.
EMV = 2000 Payoff = 500 x 4 = 2000
600
, 5%
Payoff = 500 x 6 = 3000
EMV = 0.3*1200 + 0.45*2100
+ 0.2* 3000 + 0.05*3000 =
2055
Your decision?: Self Sale
Risk Profile
Risk profile is the probability distribution for
the payoff associated with a particular action.
Payoff = 3000
60
0
Payoff = 3000
The risk profile shows all the possible economic outcomes and
provides the probability of each: it is a probability distribution for
the principal output of the model.29
Example. 6
Mary is a manager of a gadget factory. Her factory has been
quite successful the past three years. She is wondering
whether or not it is a good idea to expand her factory this
year. The cost to expand her factory is $1.5M. If she does
nothing and the economy stays good and people continue to
buy lots of gadgets she expects $3M in revenue; while only
$1M if the economy is bad.
If she expands the factory, she expects to receive $6M if
economy is good and $2M if economy is bad.
She also assumes that there is a 40% chance of a good
economy and a 60% chance of a bad economy.
(a) Draw a Decision Tree showing these choices.
Solution
40 % Chance of a Good
0.4 Economy Profit = $6M
Expand Factory
0.6
60% Chance Bad
(Cost = $1.5 M) Economy Profit = $2M
If the response is favorable, the outcome can be high sales, medium sales
or low sales with probabilities of 0.15, 0.60 and 0.25 respectively. The
respective net gains or losses for the above outcomes are $450,000,
$150,000 and -$100,000. If the result of the test marketing is not
favourable, the company can abandon the product and abandoning the
product at any stage, it results into a net gain of $30,000.
Required