MGSC 660

Math & Stat Foundations for Analytics

ASSIGNMENT # 5

INSTRUCTIONS:

1. Make sure to write down the name, and student # for each student in the group
on the cover page of the assignment.

2. This assignment counts for 15% of your final grade.

3. Late submissions will not be accepted.

4. You must complete this assignment in groups. The maximum number of

students that can be in a group is 4. Each group should submit only one
assignment.

5. Assignment submissions consist of two components:

a) Write-Up: A single Word or PDF document. This must contain your
group’s answers to all the questions. You must submit this document via
MyCourses assignment submission system.
b) Appendices: Along with the main document, you can submit your
computations in a single Excel file. This excel file must contain a
separate worksheet for each question and must also be submitted on
MyCourses as well.

6. Good luck!
 Part –1 (50%)
Question 1: Freemark Abbey Winery case (25%)

Work on the Freemark Abbey Winery case (included in the course-pack) and answer
the following questions:

a. (3%) Assuming Mr. Jaeger chooses to harvest the Riesling grapes before the
storm arrives, how much money will he make?
b. (4%) Assuming Mr. Jaeger chooses to leave the grapes on the vine, what is the
probability that the grapes will end up with botrytis, and how much money
will he make if that occurs?
c. (10%) Taking account of all the various possibilities, what should Mr. Jaeger
do?
d. (3%) How much should Mr. Jaeger be willing to pay to learn whether the
storm
really will hit the Napa Valley?
e. (5%) How much should Mr. Jaeger be willing to pay to learn whether botrytis
would form if the storm were to hit the Napa Valley?

Question 2: Revenue management on Air Canada Flight 660 (25%)

Air Canada flight 660 travels from Montreal to Toronto. The airplane assigned to the flight
is an Airbus A220 (formerly known as Bombardier CS300) having 125 seats. All of the
seats are coach class. There is no first-class section. On the other hand, the airline sells two
types of tickets discount, and full-fare. Discount tickets cost $400 each and must be
purchased at least twenty-one days ahead of time. Full-fare tickets cost $1000 each, are
refundable, and are available up to the time of departure on a space-available basis.
Virtually all full-fare tickets are sold within twenty-one days before departure.

The airline is trying to decide how many seats to set aside for full-fare passengers on its
busy Monday morning flight. The seat inventory control policy follows a nested structure.
That is, when discount tickets are offered, full-fare tickets must also be offered. A
protection level can be established to “protect” seats for full-fare customers. The protected
seats for full-fare customers are only available to full-fare customers. If too many seats are
set aside for the full-fare customers, the airplane may depart with many empty seats should
the demand be low. If too few seats are set aside, revenues again may be low due to the
number of high-paying passengers being turned away because too many discount tickets
were sold.

The airline has estimated the demand for all two types of tickets. The demand distributions
are all normal. The means and standard deviations are reported in the following table. In
addition, the demands for the two types of tickets are assumed to be independent.

Type Mean Standard deviation

Discount 120 20
 Full 30 10
To simplify the problem, assume the demand for the two types arrive sequentially with
discount first and full-fare last. Air Canada wishes to determine how many seats to set aside
for full-fare in order to maximize the expected revenue from flight 660.

(a) (15%) Construct a simulation model in Excel (or in Python) to simulate the
following policy: protect 30 seats exclusively for full-fare customers. Simulate for
1000 times.
What are the mean and standard deviation of the revenue? Calculate the 95%
confidence interval for the mean. What do you think about the policy?

(b) (5%) Use Data Table function in Excel (or create a function in Python) to optimize
the mean of the revenue by varying the protection level for full-fare customers
between 20 and 40. Report the mean and standard deviation of the revenue
corresponding to the optimal policy. Calculate the 95% confidence interval for the
mean of the revenue corresponding to the optimal policy.

(c) (5%) Find the 95% confidence interval for the probability of optimal revenue
greater than or equal to the mean revenue that corresponds to protection level of 30
seats (i.e., the mean revenue that you found in part a).
 Part 2 (50%)

In this part, we want you to develop a simulation model for the following problem and
report your findings. Format: Max. 10 pages, single-space.

Put-Call Parity

Verify “put-call parity”. Let S be the stock price, X be the strike price, T be the time to
expiration, and r be the risk-free rate. Further, let P be the value of a put option and C be
the value of a call option with those given parameters (S, X, T, and r) and any volatility.
Put-call parity asserts that:
𝑆 + 𝑃 − 𝐶 = 𝑋𝑒 −𝑟𝑇

Note that since this is assumed to hold for any market scenario, all financial risk has been
hedged away. Test if put-call parity holds by making a choice of the parameters and
estimating P and C by developing a simulation model.

Do this for several choices of parameters. Of course, the simulation results will only be
approximate by the nature of the method.

Your report should include the following sections and content described below:
1. Executive Summary (5) - Your report should start with an executive summary
that explains the problem and your findings.
2. Problem Introduction (10) - This section should lay out the problem, necessary
terminology and explain what the call and put options are.
3. Model development (15) - This section should:
a. define your parameter values that you would need to set up your
simulation model.
b. define your random variables and explain what they model.
4. Statistical Analysis (15) - This section should:
a. explain how you sample from these random variables.
b. discuss how you decide on your sample size.
5. Conclusion (5) - This section should include:
a. a discussion on how you interpret the data to draw more insightful
conclusions.
b. robust recommendations based on the data analysis.
c. a brief discussion on how your model and analysis can be extended for
further analysis.

Keep in mind the following:

• The flow of the report should be organized for the coherence of the content.
• If need be, use visual presentation of data to provide a more impactful report,
which is easier to interpret.