Search results

EuroPet S.A. was a multinational company operating gas stations in many European countries. There was a growing propensity for supermarkets to attach gas stations to their retail operations, which was developing into a major threat to EuroPet. As a result, in the mid-1990s, the company began to develop and brand its own convenience stores co-located with its gas stations. However, the company was spending much more on advertising the convenience stores than its competitors did. Management now had to decide if the increase in sales attributed to advertising efforts justified the advertising spend by analyzing the market data from one large metropolitan area: Marseille, France.

Students will learn: how to use cross-tabs and other marketing research tools to identify segmentation descriptors; how to analyze data and interpret results; and how these research results could guide new product development and positioning strategies in order to effectively target relevant customer segments.

The financial success of dairy farms depends critically on the price of their main output, milk. Large volatility in the price of milk poses a considerable business risk to dairy farms. This is particularly true for family-run dairy farms. The question then arises: how can a farm owner hedge the milk price risk? The standard approach to establish a price floor for a commodity such as milk is to purchase put options on commodity futures. At the Chicago Mercantile Exchange, farmers can buy put options on the price of a variety of milk products. However, the price a farm receives for its milk depends on many factors and is unique to the farm. Thus, a farmer cannot directly buy put options on the price he receives for the milk his farm produces. Instead the farmer needs to determine which of the options available for trade at the Chicago Mercantile Exchange offer the best hedge for his own milk price. The assignment in this case is to examine historical data on several prices of milk products and the milk price received by a family-run dairy farm in California. Students need to find the price that is most closely correlated to the farm's milk price and to then choose options with the appropriate strike price that serve as the best hedge for the farm's price risk.

The objective is to expose students to an interesting but simple finance application of linear regression analysis. To solve the case, students must run several simple linear regressions, then use the best regression model they find to make a prediction for the dependent price variable and analyze the prediction interval in order to achieve the desired objective outlined in the case. By completing the case, students will acquire a good understanding of their regression model and its usefulness.

thyssenkrupp Steel Europe, a major European steel company, operates a so-called push-pickling line (PPL) in Bochum, Germany. The PPL produces a particular type of steel strips that are sold to B2B customers, mainly in the automotive industry. In spring 2014, a senior vice president of thyssenkrupp Steel's production operations and one of his production managers notice that over the span of ten years the production facility regularly did not meet its planned production volumes. They set out to determine the drivers for the deviations from planned production figures with the ultimate goal to improve the production planning process at the Bochum PPL. Students will step into the shoes of Markus Schulze a production manager at thyssenkrupp Steel as he searches for performance drivers at the Bochum PPL and analyzes recent production data to build a forecasting model for production planning.

Jennifer McDougall is considering investing in mutual funds for the first time, and has narrowed her options down to three: one that is domiciled in Germany, and two that are domiciled in Luxembourg. As a cautious and risk-averse investor, Jennifer has done extensive research on the three funds, and has come across a curious fact: the beta of the German fund is surprisingly low. After speaking to her financial planner, she learns there is no legal requirement in Germany for mutual funds to compute net asset values at a particular time of the day. If the German fund is closing its books in the middle of the day and its net asset values reflect its midday holdings, rather than end-of-day holdings, this could explain the low beta. Thus, the German fund might appear less risky, without actually being so. Jennifer needed to get a clearer picture of what was going on before making her decision.

Using the data provided with the case, students will determine the closing time of the three funds and how that affects the beta of each. Then they must make a recommendation about which fund would be the best investment for Jennifer.

An asset management company must replace the manager of its two signature mutual funds, who is about to retire. Two candidates have been short-listed. The management team is divided and cannot decide which of the two candidates would make the better mutual fund manager. The retiring manager presents a linear regression model to examine success factors of mutual fund managers. This linear regression is the starting point for the subsequent analysis.

Application of linear regression analysis to analyze the performance of mutual fund managers.

The decision maker is in charge of procurement auctions at the department of transportation of Orangia (a fictitious U.S. state). Students are asked to assist him in estimating the winning bids in various auctions concerning highway repair jobs using data on past auctions. The decision maker is faced with various professional, statistical, and ethical dilemmas.

To analyze highway procurement auctions from the buyer-auctioneer perspective, establish basic facts regarding the project price-to-estimated cost ratio, set up and estimate a structural regression model to predict the winning bid, and compute the probability the winning price will be below estimated cost. Difficulties include heteroskedasticity, logarithmic specification, and omitted variable bias. Also to estimate a Logit regression and predict bidder collusion probability.

An angel/venture capitalist could invest in an Internet sheet-music publishing start-up. The chance of success multiplied by the value, if successful, suggests that this isn't a good investment. Nevertheless, several friends suggest the optionality present in the venture: abort an unsuccessful website and sell the technology; switch the technology if the website is good, expand, buyout. Decision trees and Monte Carlo simulations are used to value these options, which make the opportunity look very attractive.

The organizers of a music festival may use video from the Friday concert to create a DVD to sell to those who come to the Saturday concert. Attendance on Saturday is uncertain, as is the percentage of those who attend on Saturday who will buy the DVD. Is this a good project? If so, what number of DVDs should be burned early Saturday morning and offered for sale at that evening’s performance? By that time, Friday attendance is known, as well as whether it rained on Friday, and there is a forecast for whether it will rain on Saturday. Historical information on these variables may help us predict Saturday attendance using multiple regression; together with the results of a marketing survey, such analysis will help us make better purchasing decisions. This case series (see also the B case, UVA-QA-0708) can be used to illuminate a multitude of concepts that are covered in basic decision-analysis courses. The series starts by examining the role of uncertainty in decision-making, proceeds through the estimation of probability distributions from sample data with multiple regression, culminates in the development of a full decision model, and ends with a qualitative and quantitative analysis (with a tornado diagram) of how to add value and reduce risk. Key pitfalls for students are failing to recognize both limits on sales (supply and demand), incomplete reasoning in the determination of the attendance probability distribution, and oversimplifying the full forecast model.

The investment-strategy decisions of a Darden Capital Management student-portfolio management team in 2004 are examined. Case materials allow students to estimate CAPM-based expected returns using market data. The case focuses on introducing the portfolio-allocation decision; exploring the relevance of various investment-risk metrics; developing the intuition of diversification, market risk, and the capital-asset-pricing model; building judgment on how to appropriately estimate the CAPM parameters using available market data; and discussing the fundamental concepts of market efficiency.

Dr. Lawrence Frazier was an emergency room physician who was an employee of Honore Staffing Services of Baton Rouge, Louisiana. He worked at Methodist Health System hospital in Grant, Georgia. He had recently added the title of ER Medical Director and served as liaison between Honore staffing and the Methodist hospital. His additional duties included overseeing the other physicians which staff the emergency room. Methodist had a bonus system in place based on obtaining 31 patients’ satisfaction surveys each month. Dr. Frazier believed that the small sample lead to erroneous results and created problems for the physicians under his supervision. He wanted to change the data collection process (e.g. sample size collected, instrument), but encountered obstacles when he broached the subject with his hospital administrators.