This paper uses factor models to identify and estimate distributions of counterfactuals. We extend LISREL frameworks to a dynamic treatment effect setting, extending matching to account for unobserved conditioning variables. Using these models, we can identify all pairwise and joint treatment effects. We apply these methods to a model of schooling and determine the intrinsic uncertainty facing agents at the time they make their decisions about enrollment in school. Reducing uncertainty in returns raises college enrollment. We go beyond the "Veil of Ignorance" in evaluating educational policies and determine who benefits and who loses from commonly proposed educational reforms.

This paper develops two methods for estimating the effect of schooling on achievement test scores that control for the endogeneity of schooling by postulating that both schooling and test scores are generated by a common unobserved latent ability. These methods are applied to data on schooling and test scores. Estimates from the two methods are in close agreement. We find that the effects of schooling on test scores are roughly linear across schooling levels. The effects of schooling on measured test scores are slightly larger for lower latent ability levels. We find that schooling increases the AFQT score on average between 2 and 4 percentage points, roughly twice as large as the effect claimed by Herrnstein and Murray (1994) but in agreement with estimates produced by Neal and Johnson (1996) andWinship andKorenman (1997). We extend the previous literature by estimating the impact of schooling on measured test scores at various quantiles of the latent ability distribution.

We review and evaluate selection methods, a prominent class of techniques first proposed by Hedges (1984) that assess and adjust for publication bias in meta-analysis, via an extensive simulation study. Our simulation covers both restrictive settings as well as more realistic settings and proceeds across multiple metrics that assess different aspects of model performance. This evaluation is timely in light of two recently proposed approaches, the so-called p-curve and p-uniform approaches, that can be viewed as alternative implementations of the original Hedges selection method approach. We find that the p-curve and p-uniform approaches perform reasonably well but not as well as the original Hedges approach in the restrictive setting for which all three were designed. We also find they perform poorly in more realistic settings, whereas variants of the Hedges approach perform well. We conclude by urging caution in the application of selection methods: Given the idealistic model assumptions underlying selection methods and the sensitivity of population average effect size estimates to them, we advocate that selection methods should be used less for obtaining a single estimate that purports to adjust for publication bias ex post and more for sensitivity analysisthat is, exploring the range of estimates that result from assuming different forms of and severity of publication bias.

When a firm allows the return of previously purchased merchandise, it provides customers with an option that has measurable value. Whereas the option to return merchandise leads to an increase in gross revenue, it also creates additional costs. Selecting an optimal return policy requires balancing both demand and cost implications. In this paper, we develop a structural model of a consumer's decision to purchase and return an item that nests extant choice models as a special case. The model enables a firm to both measure the value to consumers of the return option and balance the costs and benefits of different return policies. We apply the model to a sample of data provided by a mail-order catalog company. We find considerable variation in the value of returns across customers and categories. When the option value is large, there are large increases in demand. For example, the option to return women's footwear is worth an average of more than $15 per purchase to customers and increases average purchase rates by more than 50%. We illustrate how the model can be used by a retailer to optimize his return policies across categories and customers.choice models, consumer behavior, decisions under uncertainty, direct marketing, e-commerce, econometric models, hierarchical Bayes analysis, latent variable models, marketing operations interface, service quality, targeting, returns

This paper provides an empirical study of entry by a Wal-Mart supercenter into a local market. Using a unique frequent-shopper database that records transactions for over 10,000 customers, we study the impact of Wal-Mart's entry on consumer purchase behavior. We develop a joint model of interpurchase time and basket size to study the impact of competitor entry on two key household decisions: store visits and in-store expenditures. The model also allows for consumer heterogeneity due to observed and unobserved factors. Results show that the incumbent supermarket lost 17% volume—amounting to a quarter million dollars in monthly revenue—following Wal-Mart's entry. Decomposing the lost sales into components attributed to store visits and in-store expenditures, we find that the majority of these losses were due to fewer store visits with a much smaller impact attributed to basket size. We also find that Wal-Mart lures some of the incumbent's best customers, and that retention of a small number of households can significantly reduce losses at the focal store. Finally, certain observed household characteristics such as distance to store, shopping behavior, and product purchase behavior are found to be useful in profiling the defectors to Wal-Mart. Implications and strategies for supermarket managers to compete with Wal-Mart are discussed.entry, retail competition, Wal-Mart supercenter, frequent-shopper data