Accounting For Decision Making III

Accounting for Decision Making III
BAC IIII, BTX III, BBF III

Decision-making under Uncertainty
 Under conditions of uncertainty, the decision-
maker is not able to assign probabilities to the
possible states of nature.
A number of decision rules to cater for
different attitudes of managers have been
formulated.
 In this section we shall look at three decision
rules,
The maximax criterion, the maximin
criterion, and the minimaxregret criterion.
Maximax Criterion
 This criterion is applied by identifying the highest

outcome (or maximum payoff) that can be obtained
for each alternative being considered.
 These maximum payoffs are compared. The
alternative that promises the highest maximum
payoff is selected.
 The decision rule applied by the decision maker is:
MAXImise MAXimum payoffs.
Steps
 Construct a payoff table (also known as payoff matrix or

conditional profit table).
 For each alternative course of action, list down the
maximum possible outcomes.
 Identify the highest outcome from the list produced in
Step 2.
 Select the alternative that yields that outcome.
 These four steps are applied below.
Example
 An investor is considering three investment alternatives known as

X, Y and Z. The payoff (benefit) from these investments depends
upon the level of economic activity in the future, and the following
estimates have been made.
 Pay off table (all figures in Shs’000):
State of Nature
Alternative X Y Z
Expansion 100 52 - 45
Stability 50 110 0
Contraction - 50 - 25 180
Which investment should the investor choose?

Maximax Criterion
 This criterion is applied by identifying the highest outcome (or maximum

payoff) that can be obtained for each alternative being considered.
 These maximum payoffs are compared. The alternative that promises the
highest maximum payoff is selected.
 The decision rule applied by the decision maker is: MAXImise MAXimum
payoffs.
 Steps
 1. Construct a payoff table (also known as payoff matrix or conditional
profit table).
 2. For each alternative course of action, list down the maximum possible
outcomes.
 3. Identify the highest outcome from the list produced in Step 2.
 Select the alternative that yields that outcome.
 4. These four steps are applied below.
Payoff Table…Example 1
 If we select investment A, the maximum payoff we can ever expect to
earn is Shs.100,000; if we select investment Y, the highest payoff
possible is Shs.110,000: and should we select investment Z, the best we
can hope for is Shs.180,000.
 Action Maximum Payoff
 Shs. ‘000
 X 100
 Y 110
 Z 180
 The highest outcome from the above list:
 The highest payoff from the above list of maximum payoffs is
Shs.180,000, promised by Investment Z.
 Select the alternative that yields that outcome
 Therefore, using maximax criterion, investment Z should be selected.
Maximin Criterion
 This criterion is applied by identifying the lowest outcome (or

minimum payoffs) that can be obtained for each alternative
being considered. These minimum payoffs are compared.
 The alternative associated with the best of these lowest
outcomes is selected. The idea is to avoid the alternative that
yields the lowest profit or one that yields the highest loss.
 The decision rule applied by the decision maker is: MAXImise
the MINimum payoffs. This criterion, when expressed in terms
of losses (or negative payoffs), is also known as minimax
criterion. The rule in this case is MINimise the MAXimum
losses.
 Steps
 1. Construct a payoff.
 2. For each alternative course of action, list down the minimum possible
outcome.
 3. Identify the highest outcome from the list produced in Step 2.
 4. Select the alternative that yields that outcome.
 Considering the previous example, minimum payoff will be:
 Action Minimum Payoff
 Shs ‘000
 X - 50
 Y - 25
 Z - 45
 Of these possible losses, the least is a loss of Shs.25,000 which could arise
if we invested in investment B and the economy turned out to be stable.
Therefore, using maximin criterion, we select investment Y.
Minimax regret
 The above two criteria make use of a payoff table or

payoff matrix. A payoff table shows the outcome for
each course of action conditional upon the occurrence
of each state of nature. The third criterion, the
minimax regret, makes use of a regret matrix or
opportunity loss matrix.
 This criterion is applied by identifying the highest
opportunity loss (or maximum regret) that can arise
for each alternative being considered. From these
regrets, the lowest is identified and the associated
course of action is selected.
 The Opportunity losses are obtained by:
 Identifying the optimal course of action for each state of nature; and
 1. Calculating, under each state of nature, the loss that would arise
from selecting a particular course of action rather than the optimal
choice.
 Steps
 1. Construct a regret matrix (also known as an opportunity loss
table).
 2. For each alternative course of action, list down the maximum
regrets.
 3. Identify the lowest regret from the list produced in Step 2.
 4. Select the alternative that generates that regret.
Constructing a regret matrix
Regret is the cost, to the decision maker, of being uncertain. If

the decision maker knew with certainty which state of nature, he
was going to obtain in the future, he/she would select the option
that yields the highest outcome under that particular state of
nature.
Based on the previous example, the following regret matrix:
State of Nature
Alternative Expansion Stability
Contraction
 X 0 60 230
 Y 48 0 205
 Z 145 110 0
Regret table
 We extract the following data:

 Action Maximum Regret
 Shs ‘000
 X 230
 Y 205
 Z 145
 The lowest regret is Shs.145,000 which is associated
with Investment Z.
 We therefore would select Investment Z, under the
Minimax regret criterion.
Decision Making Under Conditions of Risk
 Under conditions of risk, the decision-maker is able to

assign probabilities to the possible states of nature.
 The assignment of probabilities, among other things,
enables the decision-maker to calculate expected
values.
 Two decision criteria, the expected value (EV)
criterion and the expected opportunity loss (EOL)
criterion are discussed using example below.
 Example 2
 Assume the same data as in Example 1 above, with
the following additional information.
 State of Nature Probability of Occurrence
 Expansion 0.20
 Stability 0.45
 Contraction 0.35
 Solution
 Expected values for each of the three options X, Y,
and Z on the basis of the information in Example 2,
are as follows:
 EV (Shs ‘000)
X (0.2 x 100) + (0.45 x 50) + (0.35 x -50) = 25.0
Y (0.2 x 52) + (0.45 x110) + (0.35 x -25) = 51.15
Z (0.2 x -45) + (0.45 x 0) + (0.35 x 180) =
54.0
 On the basis of expected values, Investment Z is the
best, and should therefore be chosen

Accounting For Decision Making III

Uploaded by

Copyright:

Available Formats

Accounting For Decision Making III

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Accounting For Decision Making III

Uploaded by

Copyright:

Available Formats

Accounting for Decision Making III

BAC IIII, BTX III, BBF III

 This criterion is applied by identifying the highest

 Construct a payoff table (also known as payoff matrix or

 An investor is considering three investment alternatives known as

Which investment should the investor choose?

 This criterion is applied by identifying the highest outcome (or maximum

 This criterion is applied by identifying the lowest outcome (or

 The above two criteria make use of a payoff table or

Regret is the cost, to the decision maker, of being uncertain. If

 We extract the following data:

 Under conditions of risk, the decision-maker is able to

You might also like