GARP-EFM: Improving Foundation Models with Revealed Preference Structure

2026-03-25

AI Summary

This work addresses the neglect of economic rationality in existing time series foundation models for demand forecasting, which often yields predictions inconsistent with consumer utility-maximizing behavior. To bridge this gap, the authors propose GARP-EFM, a novel approach that incorporates the Generalized Axiom of Revealed Preference (GARP) as a structural constraint during synthetic data generation. Leveraging Afriat’s theorem, the method constructs GARP-consistent price-quantity time series to fine-tune the Amazon Chronos-2 model. By embedding economic theory directly into the fine-tuning process of a Transformer-based architecture, GARP-EFM achieves significant performance gains over the zero-shot Chronos-2 across all forecast horizons, demonstrating the efficacy of theory-informed synthetic data in enhancing predictive accuracy.

Technology Category

• Game Theory and Economic Paradigms: Mechanism Design
• Machine Learning: Time-Series/Data Streams
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Modern pretrained time-series foundation models can forecast without task-specific training, but they do not fully incorporate economic behavior. We show that teaching them basic economic logic improves how they predict demand using an experimental panel. We fine-tune Amazon Chronos-2, a transformer-based probabilistic time-series model, on synthetic data generated from utility-maximizing agents. We exploit Afriat's theorem, which guarantees that demand satisfies the Generalized Axiom of Revealed Preference (GARP) if and only if it can be generated by maximizing some utility function subject to a budget constraint. GARP is a simple condition to check that allows us to generate time series from a large class of utilities efficiently. The fine-tuned model serves as a rationality-constrained forecasting prior: it learns price-quantity relations from GARP-consistent synthetic histories and then uses those relations to predict the choices of real consumers. We find that fine-tuning on GARP-consistent synthetic data substantially improves prediction relative to zero-shot Chronos-2 at all forecast horizons we study. Our results show that economic theory can be used to generate structured synthetic data that improves foundation-model predictions when the theory implies observable patterns in the data.

Problem

• foundation models
• revealed preference
• demand forecasting
• economic behavior
• time-series prediction

Innovation

• GARP
• foundation models
• revealed preference
• synthetic data
• economic rationality