4 Decision Making under Risk

ZHAN, Wenjie (Professor)

School of Management, Huazhong
University of Science & Technology
Tel: 027-87556472
Email: wjzhan@mail.hust.edu.cn
 4 Decision Making Under Risk
(or Decision Making with Probabilities)

4.1 Decision Making Under Risk

4.2 Examples for Decision Making
Under Risk
 4 Decision Making Under Risk
(or Decision Making with Probabilities)
Risk refers to the situation in which the outcome
of each action is not certain, but where the
probabilities of the different states of nature
(and hence of the alternative outcomes) can be
determined.
Life is uncertain! We must deal with risk!
A probability is a numerical statement about
the likelihood that an event will occur.
 4.1 Decision Making Under Risk
(or Decision Making with Probabilities)

(1) Probability
(2) Expected Monetary Value (EMV)
(3) Expected Value of Perfect Information
(EVPI)
(4) Expected Opportunity Loss (EOL)
(5) Sensitivity analysis to probability
(6) An Example
 (1) Probability

The probability, P, of any event or state

of nature occurring is greater than or
equal to 0 and less than or equal to 1.
That is: 0 P(event) 1

The sum of the simple probabilities for

all possible outcomes of an activity
must equal 1 .
 Objective Probability
Determined by experiment or observation:
Probability of heads on coin flip
Probability of spades on drawing card from deck
Number of times event occurs
P ( event ) =
Total number of outcomes or occurrences

P(A): probability of occurrence of event A

n(A): number of times that event A occurs
n: number of independent and identical repetitions of
the experiments or observations

P(A) = limit
n
n(A) / n
 Subjective Probability

Determined by an estimate based on

personal belief,

judgment

experience,

knowledge of a situation

Probability is taken as representing the

decision makers degree (or strength) of belief
that the system will adopt a certain state.
 (2) EMV criterion
The Expected Monetary Value (EMV )
The expected monetary value (EMV ) is the
outcome anticipated when the range of pay-offs
have attached to them some estimate of objective
probability or subjective likelihood of potential
outcome.
Such probabilities may have been derived from
market research, from sales forecasting, or may
simply be a guesstimate based on some hard
evidence and experience of the decision maker.
 (2) EMV criterion
Decision making when there are several possible
states of nature and we know the probabilities
associated with each possible state.
Most popular method is to choose the alternative with
the highest expected monetary value (EMV)
EMV (alternative i) = (payoff of first state of nature)
x (probability of first state of nature)
+ (payoff of second state of nature)
x (probability of second state of nature)
+ + (payoff of last state of nature)
x (probability of last state of nature)
 (2) EMV criterionan Example
Table 1: Payoff Table for Thompson Lumber
State of Nature
Alternative Favorable Unfavorable
Market ($) Market ($)
Construct a large plant 200,000 -180,000
Construct a small plant 100,000 -20,000
Do nothing 0 0
Probabilities 0.50 0.50
 Each market has a probability of 0.50
Which alternative would give the highest EMV?
The calculations are

EMV (large plant) = (0.50)($200,000) + (0.50)($180,000)

= $10,000
EMV (small plant) = (0.50)($100,000) + (0.50)($20,000)
= $40,000
EMV (do nothing) = (0.50)($0) + (0.50)($0)
= $0

EMV (alternative i) = (payoff of first state of nature)

x (probability of first state of nature)
+ (payoff of second state of nature)
x (probability of second state of nature)
+ + (payoff of last state of nature)
x (probability of last state of nature)
 An Example: EMV for
Thompson Lumber
Table 1: Payoff Table for Thompson Lumber
STATE OF NATURE
FAVORABLE UNFAVORABLE
ALTERNATIVE MARKET ($) MARKET ($) EMV ($)
Construct a large
200,000 180,000 10,000
plant
Construct a small
100,000 20,000 40,000
plant
Do nothing 0 0 0
Probabilities 0.50 0.50
Largest EMV
 (3) Expected Value of Perfect
Information (EVPI)
EVPI places an upper bound on what you should pay for
additional information.
EVPI = EVwPI Maximum EMV
EVPI is the increase in EMV that results
from having perfect information
EVwPI is the long run average return if we have perfect
information before a decision is made.
We compute the best payoff for each state of nature since we dont
know, until after we pay, what the research will tell us
EVwPI = (best payoff for first state of nature)
x (probability of first state of nature)
+ (best payoff for second state of nature)
x (probability of second state of nature)
+ + (best payoff for last state of nature)
x (probability of last state of nature)
 (3) Expected Value of Perfect
Information (EVPI)
1. Best alternative for favorable state of nature is
build a large plant with a payoff of $200,000
Best alternative for unfavorable state of nature is
to do nothing with a payoff of $0
EVwPI = ($200,000)(0.50) + ($0)(0.50) = $100,000
2. The maximum EMV without additional
information is $40,000
EVPI = EVwPI Maximum EMV
= $100,000 - $40,000
So the maximum Thompson
= $60,000 should pay for the additional
information is $60,000
 State of Nature
Alternative Favorable Unfavorable EMV
Market ($) Market ($)
Construct a large plant 200,000 -180,000 10,000
Construct a small plant 100,000 -20,000 40,000
Do nothing 0 0 0
EVwPI =
Perfect Information 200,000 0
100,000
Compute EVwPI
The best alternative with a favorable market is to build a large
plant with a payoff of $200,000. In an unfavorable market the
choice is to do nothing with a payoff of $0
EVwPI = ($200,000)*.5 + ($0)(.5) = $100,000
Compute EVPI = EVwPI max EMV = $100,000 - $40,000 = $60,000
The most we should pay for any information is $60,000
 (3) EVPIIn-Class Example
Using the table below compute EMV, EVwPI, and
EVPI.
State of Nature
Alternative Good Average Poor
Market($) Market($) Market($)
Construct small plant 75,000 25,000 -40,000
Construct large plant 100,000 35,000 -60,000
Do nothing 0 0 0
Probability 0.25 0.50 0.25
 (3) EVPI Solution
State of Nature
Alternative Good Average Poor EMV
Market($) Market($) Market($)

Construct small plant 75,000 25,000 -40,000 21,250

Construct large plant 100,000 35,000 -60,000 27,500
Do nothing 0 0 0 0
Probability 0.25 0.50 0.25 /
EVwPI=
Perfect Information 100,000 35,000 0
42,500
 (3) EVPI Solution

EVPI = EVwPI - max(EMV)

EVwPI = $100,000*0.25 +$35,000*0.50+

+0*0.25
= $42,500

EVPI = $ 42,500 - $27,500

= $ 15,000
 (4) Expected Opportunity Loss

Expected opportunity loss (EOL) is the cost

of not picking the best solution.
First construct an opportunity loss table.For
each alternative, multiply the opportunity
loss by the probability of that loss for each
possible outcome and add these together
Minimum EOL will always result in the
same decision as maximum EMV.
Minimum EOL will always equal EVPI.
 (4) EOL: an Example
Table 1: Payoff Table for Thompson Lumber
State of Nature
Alternative Favorable Unfavorable
Market ($) Market ($)
Construct a large plant 200,000 -180,000
Construct a small plant 100,000 -20,000
Do nothing 0 0
Probabilities 0.50 0.50
 (4) EOL: an Example
Table 2: The Opportunity Loss for Thompson Lumber

STATE OF NATURE
FAVORABLE UNFAVORABLE
ALTERNATIVE MARKET ($) MARKET ($) EOL
Construct a large 200,000 -
0-(-180,000) 90,000
plant 200,000
Construct a small 200,000 -
0-(-20,000) 60,000
plant 100,000
Do nothing 200,000 - 0 0-0 100,000
Probabilities 0.50 0.50
 STATE OF NATURE
FAVORABLE UNFAVORABLE
ALTERNATIVE MARKET ($) MARKET ($) EOL
Construct a large plant 0 180,000 90,000

Construct a small plant 100,000 20,000 60,000

Do nothing 200,000 0 100,000

Probabilities 0.50 0.50 Minimum EOL

EOL (large plant)= (0.50)($0) + (0.50)($180,000) = $90,000

EOL (small plant)=(0.50)($100,000) + (0.50)($20,000) = $60,000

EOL (do nothing)= (0.50)($200,000) + (0.50)($0) = $100,000

 (4) Expected Opportunity Loss
(EOL)
The minimum EOL will always result
in the same decision (NOT value) as
the maximum EMV.

The EVPI will always equal the

minimum EOL
EVPI = minimum EOL
 (5) Sensitivity Analysis
Sensitivity analysis examines how our decision
might change with different input data.
For the Thompson Lumber example:
P = probability of a favorable market
(1 P) = probability of an unfavorable market

State of Nature
Alternative
Favorable Market ($) Unfavorable Market ($)

Construct a large plant 200,000 -180,000

Construct a small plant 100,000 -20,000
Do nothing 0 0
Probabilities P 1-P
 (5) Sensitivity Analysis
EMV(Large Plant) = $200,000P $180,000(1 P)
= $200,000P $180,000 + $180,000P
= $380,000P $180,000
EMV(Small Plant) = $100,000P $20,000(1 P)
= $100,000P $20,000 + $20,000P
= $120,000P $20,000
EMV(Do Nothing) = $0P + 0(1 P)
= $0
 (5) Sensitivity Analysis
EMV Values

$300,000

$200,000 EMV (large plant)

Point 2

$100,000 EMV (small plant)

Point 1

0 EMV (do nothing)

.167 .615 1
$100,000 Values of P

$200,000

Figure 3.1
 (5) Sensitivity Analysis

Point 1:
EMV(do nothing) = EMV(small plant)
20,000
0 = $120,000 P $20,000 P= = 0.167
120,000

Point 2:
EMV(small plant) = EMV(large plant)
$120,000 P $20,000 = $380,000 P $180,000
160,000
P= = 0.615
260,000
 (5) Sensitivity Analysis
BEST RANGE OF P
ALTERNATIVE VALUES
Do nothing Less than 0.167
EMV Values
Construct a small plant 0.167 0.615
$300,000
Construct a large plant Greater than 0.615
$200,000 EMV (large plant)
Point 2

$100,000 EMV (small plant)

Point 1

0 EMV (do nothing)

.167 .615 1
$100,000 Values of P

$200,000
Figure 3.1
 (6) Decision Making Under
Risk: an example (1/7)
Low 0.2
-15
M The EMV for the decision of M is:
Medium 0.5
10 -150.2+ 100.5+ 550.3=18.5
High 0.3
55
Low 0.2 10
BD The EMV for the decision of BD is:
Medium 0.5
25 100.2+ 250.5+ 300.3=23.5
High 0.3
30
Low 0.2
5
BA Medium 0.5 The EMV for the decision of BA is:
20 50.2+ 200.5+ 400.3=23
High 0.3
40
Fig.2-3 Completed decision tree
(pay-off and probability)
 (6) Decision Making Under
Risk: an example (2/7)
Decision

Future Probability Manufact Buy Buy

sales ure domestic abroad
level (M) (BD) (BA)
Low 0.2 -15 10 5
Medium 0.5 10 25 20
High 0.3 55 30 40
EMV 18.5 23.5 23
 Limitations of the EMV
criterion (3/7)
The EMV criterion may have been
appropriate for the decision maker
because he was only concerned with
monetary rewards, and his decision was
repeated a large number of times so
that a long-run average result would
have been of relevance to him.
 Sensitivity analysis to
probability (4/7)
Of course, the probabilities and profits used
in this problem may only be rough estimates
or, if they are based on reliable past data,
they may be subject to change.
We should therefore carry out sensitivity
analysis to determine how large a change
there would need to be in these values before
the alternative course of action would be
preferred.
 Sensitivity analysis to
probability (5/7)
Suppose the probabilities of Natures states are PL, PM, and PH. We
have:
PL+PM+ PH =1
Thus: PH =1 - PL- PM

The EMV for the decision of M is:

-15PL+ 10PM+ 55(1 - PL-PM)=55-70PL-40PM

The EMV for the decision of BD is:

10 PL + 25 PM + 30 (1 - PL-PM)=30-20PL-5PM

The EMV for the decision of BA is:

5 PL + 20 PM + 40 (1 - PL-PM)=40-35PL-20PM
 Sensitivity analysis to
probability (6/7)
For every possible PL and PM, the decision of
BD is still the best action. we should:
30-20PL-5PM > 55-70PL-40PM
30-20PL-5PM > 40-35PL-20PM
Solve the above two inequations, we have:
PL > 1/90.111
PM> 5/90.556
PL+PM >6/9 (1-PL-PM)< 3/9
PH< 3/9 0.333
It means that all the probabilities of Natures states (PL,
PM, and PH) satisfy the above conditions. The BD is still
the best action.
 The value of perfect
information (7/7)
What is the decision with perfect information?
Decision
Future Probability Manufact Buy Buy
sales ure domestic abroad
level (M) (BD) (BA)
Low 0.2 -15 10 5
Medium 0.5 10 25 20
High 0.3 55 30 40
The EMV with perfect information is : 100.2+ 250.5+ 550.3=31
EVPI=31-23.5=7.5
 4.2 Examples for Decision Making
Under Risk
Example 1: (1/6)
STATES OF NATURE
Favorable Unfavorable
ALTERNATIVES market market
Probabilites 60% 40%
Construct large plant $200,000 ($180,000)
Construct small plant $100,000 ($20,000)
Do nothing $0 $0
 Example 1: EMV Criterion (2/6)

Choose the alternative with the

maximum weighted row average.
STATES OF NATURE
Favorable Unfavorable Expected
ALTERNATIVES market market Value
Construct large plant $200,000 ($180,000) $48,000
Construct small plant $100,000 ($20,000) $52,000
Do nothing $0 $0 $0
PROBABILITIES 0.6 0.4
 Example 1: Sensitivity Analysis to
Probability (3/6)

EMV(Large Plant) = $200,000P - $180,000 (1-P)

EMV(Small Plant) = $100,000P - $20,000(1-P)

EMV(Do Nothing) = $0P + $0(1-P)

 Example 1: Sensitivity Analysis to
Probability (4/6)
250000
200000 Point 1
150000 Point 2
Small Plant
EV Values

100000
50000
0
-50000 0 0.2 0.4 0.6 0.8 1
-100000
-150000 Large Plant EMV
-200000
Values of P
 EVPI: Expected Value of Perfect
Information (5/6)
A consultant or a further analysis can aid the
decision maker by giving exact (perfect)
information about the true state: the decision
problem is no longer under risk; it will be under
certainty.
Is it worthwhile for obtaining perfectly reliable
information: is EVPI greater than the fee of the
consultant (the cost of the analysis)?
EVPI is the maximum amount a decision
maker would pay for additional
information.
 EVPI: Expected Value of Perfect
Information (6/6)
EVPI = Expected payoff with perfect information
Best expected payoff without perfect information
STATES OF NATURE
Favorable Unfavorable Expected
ALTERNATIVES market market Value
Construct large plant $200,000 ($180,000) $48,000
Construct small plant $100,000 ($20,000) $52,000
Do nothing $0 $0 $0
Example: PROBABILITIES 0.6 0.4
Expected payoff with perfect information:2000.6+00.4=120
Best expected value: 52
EVPI = 120 52 = 68
 Example 2: Tom Browns
Investment Decision (1/5)
Tom Brown has inherited $1000. He has to
decide how to invest the money for one year.

A broker has suggested five potential

investments:
Gold
Junk Bond
Growth Stock
Certificate of Deposit
Stock Option Hedge
 Example 2: The Payoff Table
(2/5)
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 States of nature

States of Nature
Alternatives Large Rise Small Rise No Change Small Fall Large Fall
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D account 60 60 60 60 60
Stock option 200 150 150 -200 -150
 Exmaple 2: The Expected Value
Criterion (3/5)
Expected Payoff = S(Probability)(Payoff)

The Expected Value Criterion Expected

Decision Large rise Small rise No change Small fall Large fall Value
Gold -100 100 200 300 0 100
Bond 250 200 150 -100 -150 130
Stock 500 250 100 -200 -600 125
C/D 60 60 60 60 60 60
Prior Prob. 0.2 0.3 0.3 0.1 0.1

EV = (0.2)(250) + (0.3)(200) + (0.3)(150) + (0.1)(-100) + (0.1)(-150) = 130

 Example 2: TOM BROWN EVPI
(4/5)

If it were known with certainty that there will be a Large Rise in the market
-100
Large rise Value of Perfect Information
The Expected
Decision 250
Large rise Small rise No change Small fall Large fall
Gold -100 100 200 300 0
Stock
Bond 500250 200 150 -100 -150
Stock
C/D
60500
60
250
60
100
60
-200
60
-600
60
Probab. 0.2 0.3 0.3 0.1 0.1

... the optimal decision would be to invest in...

Similarly,
 Example 2: TOM BROWN EVPI
(5/5)

Decision
-100
The Expected Value of Perfect Information
Large rise Small rise No change Small fall Large fall
Gold 250
-100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500500 250 100 -200 -600
C/D 60 60 60 60 60
Probab. 600.2 0.3 0.3 0.1 0.1

Expected Return with Perfect information =

ERPI = 0.2(500)+0.3(250)+0.3(200)+0.1(300)+0.1(60) = $271
Expected Return without additional information =
Expected Return of the EV criterion = $130
EVPI = ERPI - EREV = $271 - $130 = $141
 Homework 3:
An entertainment company is organizing a pop concert in London.
The company has to decide how much it should spend on publicizing
the event and three options have been identified:
Option 1: Advertise only in the music press;
Option 2: As option 1 but also advertise in the national press;

Option 3: As options 1 and 2 but also advertise on commercial radio.

For simplicity, the demand for tickets is categorized as low,

medium or high. The payoff table below shows how the profit which
the company will earn for each option depends on the level of
demand.
Demand
Option Low Medium High Profits ($000s)
1 20 20 100
2 60 20 60
3 100 60 20
 Homework 3 (cont.):
Q1: Modeling the problem with decision matrix (or pay-off
table).
Q2: Modeling the problem with decision tree.
Q3: What option the company will choose with five criteria for
the decision making under uncertainty (or pure
uncertainty):(1) The maximax criterion;(2) The maximin
criterion; (3) The Hurwicz criterion (=0.4); (4) The Laplace
criterion;(5) The minimax Regret criterion.
Q4: It is estimated that if option 1 is adopted the probabilities
of low, medium and high demand are 0.4, 0.5 and 0.1,
respectively. For option 2 the respective probabilities are 0.1,
0.3 and 0.6 while for option 3 they are 0.05, 0.15 and 0.8.
Determine the option which will lead to the highest expected
profit. Would you have any reservations about recommending
this option to the company? Ant what is the EVPI?
The end