AI@CBS

Polyhedral methods for choice-based conjoint analysis provide a means to adapt choice-based questions at the individual-respondent level and provide an alternative means to estimate partworths when there are relatively few questions per respondent as in a web-based questionnaire. However, these methods are deterministic and are susceptible to the propagation of response errors. They also assume, implicitly, a uniform prior on the partworths.

We propose and test a new approach for modeling consumer heterogeneity in conjoint estimation based on convex optimization and statistical machine learning. We develop methods both for metric and choice data. Like hierarchical Bayes (HB), our methods shrink individual-level partworth estimates towards a population mean. However, while HB samples from a posterior distribution that is influenced by exogenous parameters (the parameters of the second-stage priors), we minimize a convex loss function that depends only on endogenous parameters.

Business leaders need bold strategies to stay relevant and win. In Big Think Strategy, Schmitt shows how to bring bold thinking into your business by sourcing big ideas and executing them creatively. With the tools in this book, any leader can overcome institutionalized small think the inertia, the narrow-mindedness, and the aversion to risk that block true innovation. Your reward? Big, bold, and decidedly doable strategies that excite your employees and leave your rivals scrambling.

This paper examines the preferences of advice seekers for human information sources. We focus on the case in which advice providers can be people with high or low technical expertise (high in technical knowledge) and/or socially connected (connected to many others). Somewhat contrary to intuition, information sources who are high on social connectivity are shown to be relatively more attractive for more innovative products. Consistent with this, a meta-analysis indicates that the correlation between knowledge and opinion leadership is indeed lower for more innovative products.

Idea generation (ideation) is critical to the design and marketing of new products, to marketing strategy, and to the creation of effective advertising copy. However, there has been relatively little formal research on the underlying incentives with which to encourage participants to focus their energies on relevant and novel ideas. Several problems have been identified with traditional ideation methods. For example, participants often free ride on other participants' efforts because rewards are typically based on the group-level output of ideation sessions.

Building on the work of Dhar, Menon, and Maach (2004), this commentary describes how the compromise effect models developed in the work of Kivetz, Netzer, and Srinivasan (2004) can be extended to predict complex (business-to-business) purchase decisions and additional behavioral context effects. The authors clarify their general modeling approach and outline how it applies to choices among solutions (augmented products) and group decision making.

The authors propose and test a new "polyhedral" choice-based conjoint analysis question-design method that adapts each respondent's choice sets on the basis of previous answers by that respondent. Polyhedral "interior-point" algorithms design questions that quickly reduce the sets of partworths that are consistent with the respondent’s choices. To identify domains in which individual adaptation is promising (and domains in which it is not), the authors evaluate the performance of polyhedral choice-based conjoint analysis methods with Monte Carlo experiments.

Despite the obvious importance of understanding how business cycle fluctuations affect both individual companies and whole industries, not much marketing research focuses on the subject. Often, one only has aggregate information on the state of the national economy, even though cyclical contractions and expansions generally do not have an equal impact on every industry, nor on all firms in any given industry.