Consumer Preferences for Mixtures Using Choice Models and Experiments

Quantifying consumer preferences is a key requirement for businesses and industries to be successful in today's competitive environment. One of the key marketing research tools for that purpose is choice experiments, also known as stated preference studies. In the past few decades, much research has been done on the design of choice experiments and on the analysis of the resulting data, in marketing, but also in transportation and other sub-branches of business and economics. One product or service category that has been overlooked in all of the work done is mixtures, even though many products or services can be viewed/modelled as mixtures of ingredients. In this project, we intend to bridge the gap between choice modelling and mixture experiments, by developing and testing broad ranges of choice models tailored to mixtures and by developing an algorithmic framework to set up choice experiments for mixtures that yield a maximum of information on the consumers' preferences. We will consider various alternative models for mixture data in combination with various alternative models for choice data, hereby advancing the state of the art in the econometric modelling of preferences for a specific category of products/services. To illustrate the usefulness of our work, we will analyze two existing data sets and conduct several new choice experiments involving mixtures. For the setup of the new choice experiments, we will use our newly developed algorithmic framework.