Browse by title

We present an illustration of how marketing and structural models can be applied in a public policy context. We describe the demand model in Albuquerque and Bronnenberg (2012) to evaluate the impact of the 2009 federal policy measure known as the “Car Allowance Rebate System” program (or “Cash for Clunkers”) on prices and demand in the auto sector.

Despite the wide scope for survey evidence used in litigation, the relevance and usefulness of expert-submitted surveys in any legal context is dependent on how they are designed and implemented. The avoidance of bias in survey evidence is central to a survey’s admissibility and the probative weight accorded to the survey expert’s testimony. This chapter discusses possible sources of bias and describes methods and techniques that a survey expert can use to minimize this bias.

Bayesian econometric methods are particularly well suited for analysis of marketing data. Bayes theorem provides exact, small-sample inference within a flexible framework for assessing particular parameters and functions of parameters. We first review the basics of Bayesian analysis and examine three areas where Bayesian methods have contributed to marketing analytics – models of choice, heterogeneity, and decision theory. We conclude with a discussion of limitations and common errors in the application of Bayes theorem to marketing analytics.

The field of “Big Data” is vast and rapidly evolving. In this chapter, strict attention is paid to challenges that are associated with making statistical inferences from big data. We characterize big data by the four Vs (volume, velocity, variety and veracity) and discuss the computational challenges in marketing applications using big data. We review stochastic approximation, variational Bayes, and the methods for wide data models.

This chapter summarizes the major methods of causal inference and comments on the applicability of these methods to marketing problems.

This chapter offers an overview of Conjoint Analysis, with an eye toward implementation and practical issues. After reviewing the basic assumptions of Conjoint Analysis, I discuss issues related to implementation; data analysis and interpretation; and issues related to ecological validity. In particular, I discuss recent evidence regarding consumers’ attention in Conjoint Analysis surveys, how it may be increased and modeled, and whether responses in Conjoint Analysis surveys are predictive of real-life behavior. Each section concludes with practical recommendations.

This chapter describes an application of consumer surveys in the litigation context. This particular application of a survey differs from the typical use of market research conducted for new product development, consumer satisfaction studies, or the assessment of consumers’ willingness-to-pay for a good or service. We describe and explain why and how a survey can be an important means for either Plaintiffs or Defendants to present evidence on the interpretation of a claim (here, a so-called All Natural claim displayed on the packaging of Ben & Jerry’s ice cream), as well as to evaluate the role that such a claim can play in the consumer’s decision-making process.

We discuss the use of consumer surveys to evaluate consumer confusion in a trademark infringement case. Because trademark owners are often unable to provide evidence of actual confusion, consumer surveys can be used to evaluate the likelihood of consumer confusion over similarity of trademarks or products. We summarize the role surveys play in trademark infringement cases and discuss their use in a trademark infringement case involving artesian bottled water from the Republic of Fiji.

Consumers often “misbehave.” They save and exercise too little; they spend, eat, and drink too much and take on too much debt; they work too hard (or too little); they smoke, take drugs, have unprotected sex, and carelessly expose their private lives on social media. These misbehaviors, often characterized as time-inconsistent choices, may entail large costs not only to the individuals concerned, but also to society as a whole. In this chapter, I discuss how policy makers can take a theory-guided experimental approach, complemented by field data, to demonstrate consumer precommitment both as a revealed preference-based criterion for evaluating the need for policy intervention and as a tool for allowing consumers to limit their misbehaviors without imposing constraints on market participants’ freedom of choice.

Feature valuation is an important element of the marketing analytics toolkit and one of the primary motivations behind the popularity of conjoint analysis. We call attention to an important deficiency in current, consumer-centric, approaches. Surveys used for feature valuation need to include a reasonable competitive set. We demonstrate that equilibrium calculations are both necessary and feasible.