Probabilistic Machine Learning: New Frontiers for Modeling Consumers and their Choices

Authors

Ryan Dew, University of Pennsylvania - Marketing DepartmentFollow
Nicolas Padilla, London Business School - Department of MarketingFollow
Lan E. Luo, Columbia University - Columbia Business SchoolFollow
Shin Oblander, University of British Columbia (UBC) - Sauder School of BusinessFollow
Asim Ansari, Columbia University - Columbia Business School, MarketingFollow
Khaled Boughanmi, Cornell University - Samuel Curtis Johnson Graduate School of ManagementFollow
Michael Braun, Southern Methodist University (SMU) - Marketing DepartmentFollow
Fred M. Feinberg, University of Michigan at Ann Arbor - Marketing; University of Michigan, Stephen M. Ross School of BusinessFollow
Jia Liu, HKUST Business SchoolFollow
Thomas Otter, Goethe University Frankfurt - Department of MarketingFollow
Longxiu Tian, University of North Carolina at Chapel HillFollow
Yixin Wang, University of Michigan at Ann ArborFollow
Mingzhang Yin, University of Florida - Warrington College of Business AdministrationFollow

Publication Date

10-7-2024

Abstract

Making sense of massive, individual-level data is challenging: marketing researchers and analysts need flexible models that can accommodate rich patterns of heterogeneity and dynamics, work with and link diverse data types, and scale to modern data sizes. Practitioners also need tools that can quantify uncertainty in models and predictions of consumer behavior to inform optimal decision-making. In this paper, we demonstrate the promise of probabilistic machine learning (PML), which refers to the pairing of probabilistic modeling and machine learning methods, in pushing the frontier of combining flexibility, scalability, interpretability, and uncertainty quantification for building better models of consumers and their choices. Specifically, we overview both PML models and inference methods, and highlight their utility for addressing four common classes of marketing problems: (1) uncovering heterogeneity, (2) flexibly modeling nonlinearities and dynamics, (3) handling high-dimensional and unstructured data, and (4) addressing missingness, often via data fusion. We also discuss promising directions in enriching marketing models, reflecting recent developments in representation learning, causal inference, experimentation and decision-making, and theory-based behavioral modeling.

Document Type

Article

Keywords

machine learning, Bayesian statistics, Bayesian nonparametrics, generative models, unstructured data, representation learning, causal inference

Disciplines

Marketing

DOI

10.2139/ssrn.4790799

Source

SMU Cox: Marketing (Topic)

Language

English