Search results

IV estimation is examined when some instruments may be invalid. This is relevant because the initial just-identifying orthogonality conditions are untestable, whereas their validity is required when testing the orthogonality of additional instruments by so-called overidentification restriction tests. Moreover, these tests have limited power when samples are small, especially when instruments are weak. Distinguishing between conditional and unconditional settings, we analyze the limiting distribution of inconsistent IV and examine normal first-order asymptotic approximations to its density in finite samples. For simple classes of models we compare these approximations with their simulated empirical counterparts over almost the full parameter space. The latter is expressed in measures for: model fit, simultaneity, instrument invalidity, and instrument weakness. Our major findings are that for the accuracy of large sample asymptotic approximations instrument weakness is much more detrimental than instrument invalidity. Also, IV estimators obtained from strong but possibly invalid instruments are usually much closer to the true parameter values than those obtained from valid but weak instruments.

Intended for MBAs, this case concerns the valuation of Netflix, Inc., which was the largest U.S. online movie rental subscription service in early 2009. After reviewing Netflix's historical financial and customer relationship performance, this case presents three approaches for valuing the firm in early 2009. The first is a company-level discounted cash flow analysis based on pro forma projections of revenues, earnings, and cash flow. The second approach attempts to judge whether Netflix's prevailing market value was reasonable by comparing selected company ratios with those of comparable companies. The final approach is based on the assumption that Netflix's enterprise value (EV) was the sum of its current and future subscribers' values (discounted present values, to be exact). There is also a spreadsheet available for students (UVA-F-1610X).

Madison Kryswada, director of alumni relationships at State University, assembled a 123-school data set to explore, in her words, “the drivers of alumni giving rate. ” The alumni giving rate contributed 5% to the increasingly important U.S. News & World Report rankings of U.S. colleges and universities. Wanting to understand the relationship between this variable and school characteristics, Kryswada gave her assistant a list of four well-formed questions to answer.

Five carefully constructed problems illustrate the concepts of second-market discounting, price skimming, limit pricing, random discounting, premium pricing, and bundling.

In this case, the managers of quality assurance for a pharmaceutical company must specify a fill target for individual bottles of a new blood-pressure medicine. Higher targets lead to higher material costs but fewer seconds.

Crutchfield, a large U.S. mail-order firm specializing in consumer electronics and personal computers, must evaluate the results of a recent "prospecting" mailing to a rented list of names. A determination of the mailing requires the calculation of the lifetime value of the new customers acquired. Case data on repurchase probabilities (broken out by recency and frequency) support such a calculation. The case can also be used to introduce the RFM (recency, frequency, monetary value) framework for valuing customers.

Greg Mills describes his search for the perfect engagement ring which includes an analysis of the prices of 6,000 diamonds. An engineer, Greg hopes to impress Sarah Staggers by using regression to find an underpriced diamond. Students are asked to either select one of the 6,000 diamonds or provide point forecasts for prices of 3,142 diamonds in a hold-out sample. The instructor can use the actual prices of the held-out diamonds to evaluate student pricing models.

This case describes an innovative response-modeling project at INTUIT. The case can help students understand the basics of (and the issues surrounding) response modeling, an important tactic in data-base marketing.

The director of marketing and operations for a financial newsletter must deal with a host of issues that surround the practice of renting mailing lists and soliciting new subscribers by direct mail. The case can be used to introduce the concept and calculation of customer lifetime value.

Julia Brown, toy buyer for the Hightower Department Stores, has to decide which of three imported stuffed animals will be offered for sale by the 16 Hightower stores during the approaching Christmas shopping season. The case is appropriately used as an introduction to the concepts of least squares and regression analysis. A linear relationship between realized and test-market sales can be used to forecast the sales potential of the three proposed animals, and an explicit treatment of the uncertainty in this forecast is necessary in order to decide how many of each adopted animal to order.