Bayesian confirmatory analysis of multiple response data
Abstract
This paper proposes a Bayesian confirmatory factor-analytic probit model to reveal the latent utility structure of multiple response data commonly found in marketing surveys. It conditions model formulation on previous knowledge and imposes a parsimonious hierarchical structure involving a measurement model (to define common factors) and a structural model to explain brand choice. The confirmatory model offers some advantages over exploratory models applied to multiple response survey data. First, the model improves model identification and prediction by imposing a simpler structure that accounts for data dependencies without assuming a multivariate distribution. Secondly, using MCMC estimation, the model can easily accommodate many underlying dimensions (J) in the data, which has been challenging to address with other approaches. Lastly, the confirmatory approach offers a practical framework where the analyst has control over the specification of the latent structure of the data via informative priors. This study uniquely applies the model to test and ‘confirm’ previous knowledge and managerial hypotheses about market structures, and how brands are related and compete with one another. The study applies a fully Bayesian estimation and model choice strategy and includes a cross-validatory demarcation between test and validation sub-samples.
The full article is available to subscribers to the journal.
Author's Biography
Mauricio Ferreira is the Chief Methodologist Officer at Hypothesis Group. He leads the design and execution of the company's analytics strategy, including the development of its analytic services team and delivery of innovative custom solutions. He has extensive experience in statistical modeling and marketing analytics, particularly in areas such as consumer choice, consumer segmentation, brand equity, brand positioning, pricing, and multivariate models. Before joining Hypothesis, Mauricio was the Director/ Chief Methodologist at DRSI Group and IMS Health Consulting, and held faculty positions at Texas A&M University and Ohio University. He earned his PhD from the Ohio State University, and his work has been published in several academic journals and presented in national and international conferences.
Peter Congdon is a health statistician and quantitative geographer, and a Research Professor at Queen Mary University of London. His interests include Bayesian techniques and spatial statistics. He received his PhD in Statistics from the London School of Economics, and is author of books on Bayesian methods, most recently ‘Applied Bayesian Modelling’ (Wiley, 2014). He is an elected member of the International Statistical Institute, a Chartered Statistician and a Fellow of the Royal Statistical Society.
Yancy Edwards received his PhD in business administration (marketing) from The Ohio State University, and two graduate degrees in applied mathematics from Johns Hopkins University. His research involves building models that are more insightful and predictive of consumer behaviour. These insights are used to develop and improve product development, promotion, market segmentation, and target marketing activities. His modelling interests are in psychological and econometric models using Bayesian and Markov Chain Monte Carlo (MCMC) methods.
Citation
Ferreira, Mauricio, Congdon, Peter and Edwards, Yancy (2017, April 1). Bayesian confirmatory analysis of multiple response data. In the Applied Marketing Analytics: The Peer-Reviewed Journal, Volume 3, Issue 1. https://doi.org/10.69554/GUPT7895.
Publications LLP