MSIN0041 Marketing Science

Practice Exam 2024/25

Copyright Note to students: Copyright of this assessment brief is with UCL and the module leader(s).  If this brief draws upon work by third parties (e.g. Case Study publishers) such third parties also hold copyright. It must not be copied, reproduced, transferred, distributed, leased, licensed or shared with any other individual(s) and/or organisations, including web-based organisations, without permission of the copyright holder(s) at any point in time.

Academic Misconduct: Academic Misconduct is defined as any action or attempted action that may result in a student obtaining an unfair academic advantage. Academic misconduct includes plagiarism, self-plagiarsim, obtaining help from/sharing work with others be they individuals and/or organisations or any other form. of cheating. Definitions and penalties forAcademic Misconduct are outlined in the Academic Manual.

•     It is forbidden to:

o  Communicate or collaborate with other students or any other third parties in relation to this assessment.

o  Discuss or share assessment content with other students or third parties.

o  Copy or attempt to copy the work of another student(s).

• Such actions will be referred to the Academic Misconduct Panel, the penalties of which may include exclusion from UCL.

Referencing: In an exam no external research would normally take place and you would demonstrate your learning from the module and communicate your learning in your own words and, where appropriate, through calculations. This applies to this online assessment. If you were to make direct use of somebody else’s work, including making direct use of information generated by an artificial intelligent (AI) tool, either by memorising it or copying it word for word, you must reference and provide full citation for ALL sources used. This includes any direct quotes and paraphrased text.  If in doubt, reference it. Further guidance on referencing is to be found here: https://library-guides.ucl.ac.uk/referencing-plagiarismFailure to cite references correctly may result in your work being referred to the Academic Misconduct Panel.

Use of Artificial Intelligence (AI) Tools in your Assessment: Your module leader will explain to you if and how AI tools can be used to support your assessments. In some assessments, the use of generative AI is not permitted at all. In others, AI may be used in an assistive role which means students are permitted to use AI tools to support the development of specific skills required for the assessment as specified by the module leader. In others, the use of AI tools may be an integral component of the assessment; in these cases the assessment will provide an opportunity to demonstrate effective and responsible use of AI. See the introductory table on pages 2 and 3 of this brief to check which category use of AI falls into for this assessment. Students should refer to theUCL guidance on acknowledging use of AI and referencing AI. Failure to correctly reference use of AI in assessments may result in students being reported via the Academic Misconduct procedure. Refer to the section of the UCL Assessment success guide onEngaging with AI in your education and assessment.

IMPORTANT:

Write the question number and your chosen response on your exam answer document. There are FIVE (5) compulsory questions in total. Candidates should attempt all questions. Please keep your answers concise by keeping your answers at most 250 words per question. Be aware that a higher mark allocation does not necessarily mean a longer answer. Marks will be awarded for accuracy and relevance. Inclusion of irrelevant or incorrect information, even alongside the correct answer, may result in a deduction.

Question 1

Discuss how the choice between second-degree and third-degree price discrimination could affect a company’s product line design. What is the profit comparison between the two types of price discrimination? Do you always observe a company choose the type of price discrimination with superior predicted profits in the models learned in the lectures? If not, explain why the observation could differ from some theoretical predictions in the lectures. [20 MARKS]

Question 2

An online streaming company is trying to improve its customer retention. The company’s analytics department recently developed a new recommendation system that purportedly improves viewer engagement. To test the effectiveness of this new  system, the company applied the system to a cohort of new subscribers and monitors each user’s engagement. To benchmark the new system and improve the statistical power, the company also collected the relevant metric from a larger pool of existing subscribers with a tenure of at least 6 months.

What individual metric would you measure for engagement? How would you evaluate the new system based on the available data? Discuss the pros and cons of the approach taken by the company. [20 MARKS]

Question 3

There is a local gym charging a monthly membership fee of £30. The gym hired a marketing consultant to help it make more informed strategic decisions about its customers. Through customer survey, the consultant reported that there are segments that make up the gym’s customer base: casual goers and enthusiasts. The consultant estimated that enthusiasts made up around 30% of the customer base, whereas casual goers made up the remaining 70%.

Through a longitudinal analysis on a sample of customers, it was found that, on  average, an enthusiast member renewed the subscription with probability 80%,  whereas a casual goer member renewed the subscriptions with probability 50%.

Suppose the membership is pre-paid, that is, a gym goer needs to pay the (non-refundable) membership fee at the beginning of each month of membership. Assume the cost of acquiring a new customer is £65, which is incurred through marketing expense such as targeted advertising.

(a) Assuming the gym has a required return of 20%, can the gym afford its current customer acquisition strategy? [12 MARKS]

(b) Without any calculations, do you expected any temporal trend in the

aggregate monthly retention of a cohort of customers acquired at the same time? [8 MARKS]

Question 4

Uniqlo would like to determine who among its customer base should receive a 10%- off coupon. Recommend a data-driven approach to this question. In doing so, make sure that you identify both the type of data and the statistical methods used, as well  as how the data analysis would identify the right customers to target. [15 MARKS]

Question 5

An airline has a frequent flyer programme with two tiers of memberships: basic and premier tiers. A registered customer becomes a premier-tier member if the customer has completed at least 20 trips with the airline during the previous calendar year.

The company keeps a revenue system that keeps track of every customer’s trips with the airline.

Registered members get perks with the airline. Between the two tiers of members, the only difference between their perks is that those in the premier tier receive complimentary seat upgrades (subject to availability). The airline would like to know the effect of the seat-upgrade perk on a customer’s number of trips with the airline.

(a) Suppose the airline evaluates the effect of complimentary seat upgrades by computing the difference in the average number of booked trips between customers that have the perks and those that do not. Discuss whether this measure is biased or not. If it is biased, does it underestimate or overestimate the effect of complimentary seat upgrades? Explain your answer in the context of the question. [10 MARKS]

(b) If you are to evaluate the effect of complimentary seat upgrades, how would you measure it using existing data? Explain why your chosen estimate could provide a good measure. List and explain one limitation of your estimation strategy. [15 MARKS]