Understanding Consumer Expenditure Through Gaussian Process Choice Models

• Activity Number: 255 - Methodological Advances in Quantitative Marketing
• Type: Invited
• Date/Time: Tuesday, August 9, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Marketing
• Author(s): Alan Montgomery* and Samuel Levy
• Companies: Carnegie Mellon University and Carnegie Mellon University
• Keywords: Gaussian Process; Consumer Utility; Price Promotion; Informative Priors; Nonparametric

Abstract

Consumers change their choice as expenditures within a category increase. Traditional choice models usually make restrictive structural assumptions to specify the expenditure elasticity. This imposed functional form of utility restricts substitution patterns. Nonlinear preferences may imply consumption thresholds that could lead to improved pricing and promotion decisions. Using Gaussian process priors on utility functions, we relax the functional form of both inside goods and outside good, within the context of constrained utility maximization. We build a hierarchical model using an informative prior of a parametric functional to compensate for limited information within an individual but give the flexibility of a nonparametric form. Our model captures non-linear rates of satiation and precise baseline preferences that traditional non-homothetic parametric models fail to capture. The proposed model automatically detects non-linear patterns of consumption from the data and provide a more precise statistical inference.