Decision Making Exam 2013
Decision Making Exam 2013
Decision Making Exam 2013
School of Management
1
Question 1
A new screening test for a particular disease provides a “positive” result (i.e.
indicates that the person has the disease) for 99% of patients who have the
disease. On the other hand, the test provides a “positive” result for 1% of
patients who do not have the disease. It is known that 10% of the public have
the disease at any given time.
a). i) If a patient randomly selected from the population takes the test and
receives a “positive” result, what is the probability that they have the disease?
Use 3 decimal places.
ii) The test that is currently being used leads to a posterior probability that is
much closer to the 10% prior probability than the new test when a positive
test result for a randomly selected patient is received. Explain why this will
be.
ii) Discuss whether the EMV criterion is appropriate for the government’s
decision on whether to pay for the test.
Question 2
b) What is the rationale for using the Net Present Value (NPV) method for
appraising potential investments?
(15% of marks on question)
2
c) What are the limitations of using the NPV method, as it is traditionally
applied, with single figure (or ‘point’) estimates for the cash flows?
d) Outline the main steps involved in a carrying out a risk analysis using the
‘simulation’ method? You may use an example to explain these steps.
e) The following model is being used to estimate the first year sales (in
millions of units) of a potential new product, which has yet to be developed, in
order to decide whether it is worth continuing with its development.
There is, however uncertainty about the future value of the three factors that it
is thought will determine sales. For example, the quality achieved will depend
on whether technical problems associated with the manufacturing process will
be overcome, while the number of sales people who could be appointed with
sufficient technical knowledge to sell the product is also uncertain, as is the
ease of use of the final design that will be achieved.
3
You are given a three random numbers which could be used in the first run of
your simulation to estimate the product’s first year sales. The random
numbers are: 29, 67, 43.
Question 3
a) Describe how you might elicit the utility function for the decision makers in
this problem. Come up with example utilities for all the possible outcomes.
c) Discuss potential problems with the methodology utilised in parts (a) and
(b) in examining such a decision problem.
iii) Discuss the role that sensitivity analysis could play in this decision
Question 4
4
For each of the eight situations described below, identify why the judgments
of the managers involved may be subject to biases. (Each of the eight parts
carries an equal mark).
5
6. Only 25 % of the products launched onto the market by a company are
economic successes and survive in the market for more than a year. The
chances of any product being successful are thought to be independent of
what happens to other products. Remarkably, the last six products developed
by the company have been successful. One manager points out that the
chances of a run of success like this are about 2 in 10000. His colleague now
has to assess the probability that the latest product, which is due to be
launched next month, will be successful and hence will still be on the market a
year after its launch.
7 A company has been losing customers to a rival for several years and has
to decide whether to undertake a rebranding of its products in an attempt to
stop the decline. It is thought that, if nothing is done, another 3000 customers
will be lost this year. However rebranding is a risky strategy and it is estimated
that it will have a 30% chance of ensuring that no further customers are lost
this year, but a 70% chance that it will worsen the situation and lead to the
loss of 6000 customers.
Question 5