Knowledge that Transforms

In this paper, we provide theoretical arguments and empirical evidence for how Genetic Algorithms (GA) can be used for efficient estimation of macro-level diffusion models. Using simulations we find that GA and Sequential Search-Based-Nonlinear Least Squares (SSB-NLS) provide comparable parameter estimates when the data including peak sales are being used, for a range of error variances, and true parameter values commonly encountered in the literature. From empirical analyses we find that the forecasting performance of the GA estimates is better than that of SSB-NLS, Augmented Filter, Hierarchical Bayes, and Kalman Filter when only pre-peak sales data is available for estimation. When sales data until the peak time period are available for estimation, SSB-NLS is able to obtain parameter estimates when the starting values provided are the estimates from using GA. The estimates from GA are not biased and do not change in a systematic fashion when post-peak sales data are used, whereas the estimates from SSB-NLS are biased and change in a systematic fashion. Summarizing, we find that GA may be better suited for diffusion model estimation under the three conditions where SSB-NLS has been found to have problems.

There are many critical concerns (including the accounting for endogeneity) when one is properly estimating response functions. However, it is sometimes (certainly not always) better to leave some variables exogenous when building mathematical models intended to help decision makers. The exogenous variables allow the decision maker to better adapt the mathematical model to different situations and to incorporate myriad variables and constraints outside of the model.

The issue of delegating pricing responsibility to the salesforce is of interest to marketing academics and practitioners. It has been shown by Lal (1986) that under certain situations with information asymmetry, it is more profitable for the firm to delegate pricing authority to the salesforce than to have centralized pricing. In this paper we re-examine situations where information asymmetry exists and analyze its effect on the decision of the firm to set a price or delegate pricing responsibility to the salesforce. Using contract theory, we find that when the salesperson's private information can be revealed to the firm through contracting, centralized pricing performs at least as well as price delegation. We derive the optimal centralized pricing contract under a set of standard assumptions used in the economics and business literature.

Academic research in marketing often and rightfully tends to either build on well-established past research topics or follow well-established practices in industry. However, as technology advances, it might be possible to foresee some more enduring trends and focus research on future issues rather than on past issues. One approach would be to study emerging technologies with rapidly declining costs. Each of these emerging technologies spawns myriad applications that have the potential to dramatically impact existing markets. Interesting research topics include the study of the impact of these applications on different market participants (e.g., final consumers, the seller, the seller of complementary services, intermediaries, information providers, competitors, other industries). Research topics also include the optimal structure for products and services, given these new applications, as well as which intermediary should offer particular services. Research topics also include the interactive ability to rapidly customize marketing strategy by identifying individuals at particular points in time and under particular demand conditions. Five of these technologies include enhanced search services, biometrics and smart cards, enhanced computational speed, M-commerce, and GPS tracking.

When selling a product line, a firm has to consider the costs of communicating about the different products to the consumers. This may affect the product line design in general, and which products or services are offered in particular. The problem is that firms have to communicate to consumers, possibly through advertising, to make them consider buying the products that firms are selling. This results in the firm offering a smaller number of products than is optimal when advertising has no costs. This effect is greater the extent of consumer confusion about the advertising messages, and is reduced by a greater ability to target advertising. When offering vertically differentiated products (second-degree price discrimination), under general conditions it is optimal to advertise so that one has a greater proportion of sales of a lower-quality product than if advertising had no cost. This situation also allows the firm to charge a lower price for the high-quality product and offer a higher quality of the low-quality product than it would if advertising were without cost.

Although negotiating over prices with sellers is common in many markets such as automobiles, furniture, services, consumer electronics, etc., it is not clear how a haggling price policy can help a firm gain a strategic advantage or whether it is even sustainable in a competitive market. In this paper, we explore the implications of haggling and fixed prices as pricing policies in a competitive market. We develop a model in which two competing retailers choose between offering either a fixed price or haggling over prices with customers. There are two consumer segments in our analysis. One segment, the hagglers, has a lower opportunity cost of time and a lower haggling cost than the other segment, the nonhagglers. When both retailers follow the same pricing policy, then a haggling policy is more profitable than a fixed-price policy only when the proportion of nonhagglers is sufficiently high. We find two kinds of prisoners' dilemma: under some conditions, a more profitable haggling policy can be broken by a fixed-price policy, and under other conditions, a fixed-price policy can be broken by a haggling policy. Surprisingly, we show that under some conditions, an asymmetric outcome with one retailer haggling and the other offering a fixed price is also an equilibrium.

The main objective of this paper is to provide a decision-support system of micro-level customized promotions, primarily for use in online stores. Our proposed approach utilizes the one-on-one and interactive nature of the Internet shopping environment and provides recommendations on when to promote how much to whom. We address the issue by first constructing a joint purchase incidence-brand choice-purchase quantity model that incorporates how variety-seeking/inertia tendency differs among households and change over time for the same household. Based on the model, we develop an optimization procedure to derive the optimal amount of price discount for each household on each shopping trip. We demonstrate that the proposed customization method could greatly improve the effectiveness of current promotion practices, and discuss the implications for retailers and consumer packaged goods companies in the age of Internet technology.

We develop a brand choice model with learning based on the Kalman filter methodology. The model enables us to separate the effects of contemporaneous marketing promotions from the impact of the perceived quality valuation accrued through product usage over time. We also account for idiosyncratic consumer learning and preferences. The results point to the presence of heterogeneity in the valuation carryover coefficients across consumers and brands. In contrast to our expectations, a higher price is not important for most of the consumers in the sample. The model enables us to compare brands in terms of their memorability, which determines brand salience on the next purchase occasion. Our findings suggest that price promotions may be deficient as a tool to increase market share in the studied product category. The proposed model is applicable to other consumer goods contingent on consumers' being sufficiently motivated to learn their own preferences via personal experience. Brand managers can use the model for comparative diagnostics and market performance simulation under different price and promotion scenarios. This paper is instructive to the application of a relatively new methodology; we illustrate the analytical potential of the model by demonstrating its inferential power in a specific marketing context.

This paper examines a market where the provision of information service is costly, but information service has the characteristics of a public good. Consumers, on the other hand, can use the information service to make an informed purchase decision and derive higher utility from consuming their ideal product. However, after receiving the information service from an information service provider, consumers can easily free ride by purchasing at low-price sellers who do not provide any information service. The paper examines the competition where sellers compete by providing information service for horizontally differentiated products and where technology reduces consumers' search cost. It is found that in this market a seller needs to establish itself as an information service provider in order to make positive profits, even when there is free riding. A seller, however, cannot make positive profits by free riding all the time. Also, with an increase in competition in the information service market, sellers have reduced incentives to provide information service. It is also found that in this market a decrease in search cost may increase or decrease social welfare.

This paper examines four alternative product strategies available to an innovating firm in markets with network effects: single-product monopoly, technology licensing, product-line extension, and a combination of licensing and product-line extension. We address three questions. First, what factors affect the attractiveness of each of the four product strategies? Second, under what conditions will any particular strategy dominate the others? Third, what is the impact of licensing fees on the profitability of a licensing strategy? We show that offering a product line utilizes consumer heterogeneity to increase the total user base and is superior to free licensing when the innovator's cost of producing a low-quality product is low and network effects are weak. However, because of the advantage of licensing in generating a larger installed base, free licensing can dominate line extension when network effects are strong, even if the innovator suffers no cost disadvantage compared to the competitor. We also show that paid licensing trumps free licensing when the clone product has a high quality or a low cost, regardless of network effect. Finally, strong network effects make a lump-sum fee more profitable than a royalty fee (or a combination of both) because a royalty fee reduces the licensee's production.