STATISTICAL DECISION THEORY

1. A baker produces a certain type of special pastry at a total average cost of

Rs. 3 and sell it at a price of Rs. 5. This pastry is produced over the weekend and is sold during the
following week; such pastry being produced but not sold during a week’s time are totally spoiled and
have toe thrown. According to past experience the weekly demand for these pastries is never less than
78 or greater than 80. You are required to formulate action space, state of nature, payoff table and loss
table. Also suggest the suitable action based on a) Maximax, b) maximin, c) minimax d) Laplace criteria.

2. A toy company is bringing out a new type of toy. The company is attempting to decide whether to bring
out full, partial, or minimal product line, the company has three levels of product acceptance.
Management will make its decision on the basis of expected profit from the first year of production. The
relevant data are shown in the following table.

Product Anticipated Ist year profit (Rs.000’s)

Acceptance Product line

Full partial minimal

Good 80 70 50

Fair 50 45 40

Poor -25 -10 0

Take optimal decision under each of the following decision criteria

a) Maximax, b) maximin, c)minimax d) Laplace

3. A businessman wants to construct a hotel. He usually builds 25, 50 or 100 bed hotel, depending on
whether anticipated demand is low, medium or high. The businessman has been able to find out net
profits which are expressed in the table below and the corresponding probabilities are also given below.
Courses of action
Build 25 bed Build 50 bed build 100 bed
Hotel hotel hotel

States of low
nature demand 20,000 -10,000 -30,000

Medium
Demand 25,000 25,000 -5,000

High
Demand 30,000 50,000 60,000
a) Compute EMV, EPPI and EVPI
b) A research firm agrees to conduct a survey for Rs. 8,000 and provide him with information regarding the
states of nature. Should the survey be conducted?
4. A baker makes a certain kind of pastry at night and sells it the next day. It is perishable and must be
thrown if not sold during the day.
The unit cost and price of the pastry are Re.1 and Rs. 3 respectively. According to the past experience
the daily demand (in hundred ) and respective probabilities are
Demand : 20 21 22 23 24
Probability : 0.1 0.2 0.3 0.3 0.1

a) Construct the payoff table

b) Construct loss table
c) Determine the maximin and Maximax action
d) Compute the highest expected payoff with perfect information.

5. Given the following payoff matrix

-- --------------------------------------------------------------------------------------------------------
State of nature probability (Decision Rs)
expand expand expand
100 units 200 units 400 units
----------------------------------------------------------------------------------------------------------
High demand 0.4 2500 3500 5000

Medium demand 0.4 2500 3500 1000

Low demand 0.2 2500 1500 1000

----------------------------------------------------------------------------------------------------------
What should be the decision under EMV criterion. Also find EPPI and EVPI

6. A food products company is contemplating the introduction of a revolutionary new product with new
packaging to replace the existing product at much price(A1) or a moderate change in the composition of
the existing product with a new packaging at a small increase in price (A2) or a small change in the
composition of the existing except the word ‘New’ with a negligible increase in price (A3). The three
possible states of nature of events are (i) high increase in sales (ii) no change in sales (iii) decrease in
sales. The marketing department of the company worked out the payoffs in terms of yearly net profits
for each course of the action for these events (expected sales). This is represented in the following table.

---------------------------------------------------------------------------------------------------------
States of Nature Courses of action
A1 A2 A3
 --------------------------------------------------------------------------------------------------------
S1 7,00,000 5,00,000 3,00,000

S2 3,00,000 4,50,000 3,00,000

S3 1,50,000 0 3,00,000
----------------------------------------------------------------------------------------------------------

Which strategy should the company choose on the basis of a) Maximax, b) maximin, c) minimax d)
Laplace criteria.

7. An electrical manufacturing company has seen its business expand to the point where it needs to
increase production beyond its existing capacity. It has narrowed the alternatives to two approaches to
increase the maximum production capacity: (a) expansion at a cost of Rs. 8 million or (b) modernization
at a cost of Rs. 5 million. Both approaches would require the same amount of time for implementation.
Management believes that over the required payback period, demand will either be high or moderate.
The probability of high demand has been set at 0.35. If the demand is high, expansion would gross an
estimated additional Rs. 12 million but modernization only an additional Rs. 6 million, due to lower
maximum production capability. On the other hand if the demand is, moderate, the comparable figures
would be Rs. 7 million for expansion and Rs. 5 million for modernization.

a) Calculate the conditional profit in relation to various action and outcome combinations and states of
nature.
b) If the company wishes to maximize its expected monetary value, then should it modernize or expand?
c) Calculate EVPI
d) Construct the conditional opportunity loss table and also calculate EOL

8. A certain piece of equipment is to be purchased for a construction project at a remote location. This
equipment contains an expensive part which is subject to random failure. Spares of this part can be
purchased at the same time the equipment is purchased. Their unit cost is Rs. 1,500 and they have no
scrap value. If the part fails on the job and no spare is available, the part will have to be manufactured
on a special-order basis. If this is required, the total, including down time of the equipment is estimated
as Rs. 9,000 for each such occurrence. Based on previous experience with similar parts, the following
probability estimates of the number of failures expected over the duration of the project are provided as
given below.

Failure : 0 1 2 3 & above

Probability : 0.80 0.15 0.05 0.00

a) Determine optimal EMV and optimal number of spares to purchase initially.

b) Based on opportunity losses, determine the optimal course of action and optimal value of EOL
c) Determine the expected profit with perfect information and expected value of perfect information.

DECISION TREE ANALYSIS

9. A glass factory specializing in crystal is developing a substantial backlog and the firm’s management is
considering three courses of action(A1) arrange subcontracting, (A2) begin overtime production and (A3)
 construct new facilities. The correct choice depends largely upon future demand which may e low,
medium, or high. By consensus, management ranks the respective probabilities as 0.10,0.50 and 0.40. A
cost analysis reveals the effect upon the profits that is shown in the table below.

------------------------------------------------------------------------------------------------------
Profit Courses of Action
(Rs.’000) --------------------------------------------------------------------------------
If demand is A1 A2 A3
----------------------------------------------------------------------------------------------------

Low 10 -20 -150

Medium 50 60 20

High 50 100 200

----------------------------------------------------------------------------------------------------

Show this decision situation in the form of a decision tree and indicate the most preferred decision and
corresponding expected value.

10. Raman industries Ltd, has a new product which they expect has great potential. At the moment they
have two courses of action open to them

A1: To test market.

A2: To drop the product.

If they test it, it will cost Rs. 50,000 and the response could be positive or negative with probabilities of
0.70 and 0.30, respectively. If it is positive, they could either market it with full scale or drop the
product. If they market with full scale, then the result might be low, medium , or high demand, and the
respective net payoffs would be – Rs. 1,00,000, Rs. 1,00,000 or Rs. 5,00,000. These outcomes have
probabilities of 0.25, 0.55 and 0.20 respectively. If the result of the test marketing is negative they have
decided to drop the product. If, at any point, they drop the product there is a net gain of Rs. 25,000 from
the sale of scrap. All financial values have been discounted to the present.

Draw a decision tree for the problem and indicate the most preferred decision.

11. A manufacturing company has just developed new product. Based on past experience, a product such as
this will either be successful, with an expected gross return of Rs.1,00,000, or unsuccessful, with an
expected gross return of Rs.20,000. Similar products manufactured by the company have a record of
being successful about 50% of the time. The production and marketing costs of the new product are
expected to be Rs. 50,000.

The company is considering whether to market this new product or to drop it. Before making its
decision, however, a test marketing effort can be conducted at a cost of Rs. 10,000. Based on the past
experience, test marketing results have been favorable about 70 % of the time. Furthermore, products
favorably tested have been successful 80 % of the time. However, when the test marketing result has
been unfavorable, the product has only been successful 30 % of the time. What course of action should
the company pursue?
12. A businessman has two independent investments A and B available to him, but he lacks the capital to
undertake both of them simultaneously. He can choose to take A first and then stop, or if A is successful
then take B, or vice-versa. The probability of success of A is 0.7, while for B it is o.4. Both investments
require an initial capital outlay of Rs.2000 and both returns nothing if the venture is unsuccessful.
Successful completion of A will return Rs. 3,000 (over cost) and successful completion of B will return Rs.
5,000(over cost). Draw the decision tree and determine the best strategy.

13. There is a 40 % chance that a patient admitted to the hospital, is suffering from cancer. A doctor must
decide whether a serious operation should be performed or not. If the patient is suffering from cancer
and the serious operation is performed, the chance that he will recover is 70 %, otherwise it is 35 %. On
the other hand if the patient is not suffering from cancer and the serious operation is performed, the
chance that he will recover is 20 %, otherwise it is 100 %. Assume that recovery death are the only
possible results.

Construct an appropriate decision tree. What decision should the doctor take?

14. An oil company has recently acquired rights in a certain area to conduct surveys and test drillings to lead
to lifting oil if it is found in commercially exploitable quantities.
The area is considered to have good potential for finding oil in commercial quantities. At the outset, the
company has the choice to conduct further geological test or to carry out a drilling programme
immediately. On the known conditions, the company estimates that there is a 70:30 chance of further
tests showing a ‘success’.
Whether the tests show the possibility of ultimate success or not or even if no tests are undertaken at
all, the company could still pursue its drilling programme or alternatively consider selling its rights to
drill in the area. Thereafter, however, if it carries out the drilling programme, the likelihood of final
success or failure is considered dependent on the foregoing stages. Thus:
 If ‘successful’ tests have been carried out, the expectation of success in drilling is given as 80:20.
 If the tests indicate ‘failure’, then the expectation of success in drilling is given as 20:80.
 If no tests have been caried out at all, the expectation of success in drilling is given as 55:45.
Costs and revenues have been estimated for all possible outcomes and the net present value of each is
as follows.

Outcome Net present Value (Rs. million)

Success With prior tests 100
Without prior tests 120

Failure With prior tests - 50

Without prior tests - 40

Sale of exploitation rights:

Prior test show ‘success’ 65

Prior test show ‘failure’ 15
Without prior tests 45

(a) Draw up a decision (Probability) tree diagram to represent the above information; and
(b) Evaluate the tree to advise the management of the company on its best course of action.
 Bayesian Decision Rule: Posterior Analysis

15. A car hire firm keeps 5, 10 or 20 cars depending on whether the anticipated demand is low, medium, or
high. The firm knows the profit or loss which these varying demand situations given it every day. The
firm has also probability distribution of the various states of nature (demand) based on experience. The
pay-off table and the probability distribution are given below.

PAYOFF MATRIX
--------------------------------------------------------------------------------------------------
Event Strategy
(5 cars) (10 cars) (20 cars)
---------------------------------------------------------------------------------------------------
Low demand 200 -100 -300
Moderate demand 250 300 500
High demand 300 500 600
----------------------------------------------------------------------------------------------------

(PRIOR)PROBABILITUY DISTRIBUTION
-----------------------------------------------------------------------------------------------------
State of Nature S1 S2 S3
(demand) low moderate high
Probability 0.2 0.3 0.5
----------------------------------------------------------------------------------------------------
1. If the firm wishes to maximize its expected profit, what will be its optimum
decision?
2. If a firm is prepared to gather additional information for Rs. 100, should the
information be collected
3. If the firm collects additional information and gives the following conditional
probabilities for the state of nature, calculate the posterior probabilities and the
posterior pay-off table.

PAY-OFF MATRIX
-----------------------------------------------------------------------------------------------------
Actual demand Conditional probability
Demand prediction
Low moderate High
----------------------------------------------------------------------------------------------------
Low 0.60 0.30 0.10
Moderate 0.20 0.60 0.20
High 0.10 0.20 0.70
----------------------------------------------------------------------------------------------------

16. Suppose that a firm is in the process of installing a computer system for providing information
services, like preparation of payrolls, balance sheet etc., to the firms in an industrial area. The firm must
take a decision as to what size of computer system should be purchased. The three alternatives open to
the firm are: large-sized(A1), medium-sized(A2) and small-sized(A3) computer systems. The
management of the firm feels that the overall level of acceptance of its services would be either high or
low. The payoffs expected under various event-action combinations together with the estimated
probabilities of the likely demand are.

Event Prob. Actions (Aj)

A1 A2 A3

High acceptance 0.2 500,000 250,000 120,000

Low acceptance 0.8 -60,000 50,000 100,000

With this information, there are two strategies before the management of the firm:
First, to decide on the size of the computer system based on the prior information only. Second, to
conduct a market research and take decision about the size of the system based on both: the prior
information and on the information obtained from the research. Help it.

Solution:
The management must decide on its strategy – by considering relative merits of the two
strategies in terms of expected payoffs associated with them. The evaluation of the former is
the same as the analysis that we have considered so far. It is called prior analysis because it
uses only prior probabilities. On the other hand, the latter involves incorporating the
information obtained through the marketing research into the analysis and take the decision
thereafter. This is termed as the posterior analysis. We consider them in turn now.

(refer to pages 693 and 694)