Knowledge that Transforms

Random effects or mixed logit models are often used to model differences in consumer preferences. Data from choice experiments are needed to estimate the mean vector and the variances of the multivariate heterogeneity distribution involved. In this paper, an efficient algorithm is proposed to construct semi-Bayesian D-optimal mixed logit designs that take into account the uncertainty about the mean vector of the distribution. These designs are compared to locally D-optimal mixed logit designs, Bayesian and locally D-optimal designs for the multinomial logit model and to nearly orthogonal designs (Sawtooth (CBC)) for a wide range of parameter values. It is found that the semi-Bayesian mixed logit designs outperform the competing designs not only in terms of estimation efficiency but also in terms of prediction accuracy. In particular, it is shown that assuming large prior values for the variance parameters for constructing semi-Bayesian mixed logit designs is most robust against the misspecification of the prior mean vector. In addition, the semi-Bayesian mixed logit designs are compared to the fully Bayesian mixed logit designs, which take also into account the uncertainty about the variances in the heterogeneity distribution and which can be constructed only using prohibitively large computing power. The differences in estimation and prediction accuracy turn out to be rather small in most cases, which indicates that the semi-Bayesian approach is currently the most appropriate one if one needs to estimate mixed logit models.

We analyze firms' decisions to invest in customer relationship management (CRM) initiatives such as acquisition and retention in a competitive context, a topic largely ignored in past CRM research. We characterize each customer by her intrinsic preference towards each firm, the contribution margin she generates for each firm, and her responsiveness to each firm's retention and acquisition efforts. We show that a firm should invest most heavily in retaining those customers that exhibit moderate responsiveness to its CRM efforts. Further, a firm should most aggressively seek to attract those customers that exhibit moderate responsiveness to their provider's CRM efforts and those that are moderately profitable for their current provider. Investing more in customers that are more responsive does not always lead to higher firm profits, because stronger competition for such customers tends to erode the effects of higher CRM efforts of an individual firm. When firms develop a customer relationship over time to generate higher contribution margin or customer responsiveness, we show that such developments may not always be desirable, because sometimes these future benefits may lead to more intense competition and hence lower profits for both firms.

In their purchase decisions, online customers seek to improve decision quality while limiting search efforts. In practice, many merchants have understood the importance of helping customers in the decision-making process and provide online decision aids to their visitors. In this paper, we show how preference models which are common in conjoint analysis can be leveraged to design a questionnaire-based decision aid that elicits customers' preferences based on simple demographics, product usage, and self-reported preference questions. Such a system can offer relevant recommendations quickly and with minimal customer input. We compare three algorithms—cluster classification, Bayesian treed regression, and stepwise componential regression—to develop an optimal sequence of questions and predict online visitors' preferences. In an empirical study, stepwise componential regression, relying on many fewer and easier-to-answer questions, achieved predictive accuracy equivalent to a traditional conjoint approach.

When demand is uncertain, manufacturers and retailers often have private information on future demand, and such information asymmetry impacts strategic interaction in distribution channels. In this paper, we investigate a channel consisting of a manufacturer and a downstream retailer facing a product market characterized by short product life, uncertain demand, and price rigidity. Assuming the firms have asymmetric information about the demand volatility, we examine the potential benefits of sharing information and contracts that facilitate such cooperation. We conclude that under a wholesale price regime, information sharing might not improve channel profits when the retailer underestimates the demand volatility but the manufacturer does not. Although information sharing is always beneficial under a two-part tariff regime, it is in general not sufficient to achieve sharing, and additional contractual arrangements are necessary. The contract types we consider to facilitate sharing are profit sharing and buyback contracts.

Existing literature discusses a number of possible pioneering cost advantages and disadvantages. In this paper, we empirically test three different sources of long-term pioneering cost advantage—experience curve effects, preemption of input factors, and preemption of ideal market space—and three different sources of pioneering cost disadvantage—imitation, vintage effects, and demand orientation. We disentangle these sources by breaking total cost of a business unit into three different components—purchasing, production, and selling, general, and administrative (SG&A) costs—and identifying conditions that intensify or reduce the effect of the proposed source. Using two samples of business units, one for consumer goods and one for industrial goods, we find support for five of the six sources of pioneering cost advantage and disadvantage in both samples, while the advantage due to preemption of ideal market space is limited to the consumer goods sample. The unconditional analysis shows a pioneering purchasing cost advantage but even larger pioneering production and SG&A cost disadvantages. The complexity of our obtained findings suggests that managers need to think carefully about their particular conditions before making assumptions about the cost and, therefore, profit implications of a pioneering strategy.

We model the commercial World Wide Web as a directed graph that emerges as the equilibrium of a game in which utility maximizing websites purchase (advertising) in-links from each other while also setting the price of these links. In equilibrium, higher content sites tend to purchase more advertising links (mirroring the Dorfman-Steiner rule) while selling less advertising links themselves. As such, there seems to be specialization across sites in revenue models: high content sites tend to earn revenue from the sales of content, whereas low content ones earn revenue from the sales of traffic (advertising). In an extension, we also allow sites to establish (reference) out-links to each other and find that there is a general tendency to establish reference links to sites with higher content. Finally, we explore network formation in the presence of search engines and find that the higher the proportion of people using them, the more sites have an incentive to specialize in certain content areas. Our results have interesting practical implications for search-engine optimization, the pricing of online advertising, and the choice of Internet business models. They also shed light on why Google can use the web's link structure to rank sites by content.

Two models of competition between high-end and low-end products benefiting the high-end firms are presented. One is a quantity competition model, and the other is a price competition model with product differentiation. The key factor is the existence of two heterogeneous consumer groups: those who demand only high-end (name-brand) products and those who care little whether products are high or low end. We show that, under certain conditions, the profits of firms in the high-end market are larger when there are firms producing low-end products than when there are not. The existence of price-sensitive consumers who care little about product quality intensifies competition among the high-end firms. The existence of low-end firms functions as a credible threat, which induces the high-end firms not to overproduce because price-sensitive consumers buy products from the low-end firms. The result provides a new theoretical mechanism concerning the profitability and pricing of national brand firms after the entry of private labels. It has an implication for pricing and marketing strategies: Established firms should not decrease their prices after the entry of nonestablished firms.

This paper develops a model that extends the traditional aggregate discrete-choice-based demand model (e.g. Berry et al. 1995) to account for varying levels of product availability. In cases where not all products are available at every consumer shopping trip, the observed market share is a convolution of two factors: consumer preferences and the availability of the product in stores. Failing to account for the varying degree of availability would produce incorrect estimates of the demand parameters. The proposed model uses information on aggregate availability to simulate the potential assortments that consumers may face in a given shopping trip. The model parameters are estimated by simulating potential product assortment vectors by drawing multivariate Bernoulli vectors consistent with the observed aggregate level of availability. The model is applied to the UK chocolate confectionery market, focusing on the convenience store channel. We compare the parameter estimates to those obtained from not accounting for varying availability and analyze some of the substantive implications.

There is substantial literature documenting the presence of state-dependent utility with packaged goods data. Typically, a form of brand loyalty is detected whereby there is a higher probability of purchasing the same brand as has been purchased in the recent past. The economic significance of the measured loyalty remains an open question. We consider the category pricing problem and demonstrate that the presence of loyalty materially affects optimal pricing. The prices of higher quality products decline relative to those of lower quality when loyalty is introduced into the model. Given the well-known problems with the confounding of state dependence and consumer heterogeneity, loyalty must be measured in a model which allows for an unknown and possibly highly nonnormal distribution of heterogeneity. We implement a highly flexible model of heterogeneity using multivariate mixtures of normals in a hierarchical choice model. We use an Euler equations approach to the solution of the dynamic pricing problem which allows us to consider a very large number of consumer types.

We develop a robust model for estimating, tracking, and managing brand equity for multicategory brands based on customer survey and financial measures. This model has two components: (1) offering value (computed from discounted cash flow analysis) and (2) relative brand importance (computed from brand choice models such as multinomial logit, heteroscedastic extreme value, and mixed logit). We apply this model to estimate the brand equity of Allstate—a leading insurance company—and its leading competitor, which compete in multiple categories. The model captures the brand's spillover effects from one category to another. In addition, we identify the dimensions that drive a brand's image, examine the relationships among advertising, brand equity, and shareholder value, and build a decision support simulator for the focal brand. Our model provides reliable estimates of brand equity, and our results show that advertising has a strong long-term positive influence on brand equity, which is significantly positively related to shareholder value. The model, the brand equity estimates, and the decision support simulator are used by key executives across multiple functional areas and have enabled the company to substantially gain by reallocating its advertising resources to improve brand equity and shareholder value, and by offering better guidance to analysts and investors.