Search papers, labs, and topics across Lattice.
This study investigates the application of tabular foundation models (TFMs) to discrete choice estimation, a key framework in marketing and operations. The authors identify a structural limitation in TFMs due to their assumption of row-independent observations, which does not align with the set-valued nature of discrete choice data. By reformulating the approach to incorporate choice-set dependence and individual heterogeneity, they achieve an 8% improvement in predictive accuracy over hierarchical Bayesian estimation while significantly reducing computational time, particularly benefiting medium-data scenarios.
Encoding individual consumer preferences in tabular foundation models can outperform traditional Bayesian methods by 8% in predictive accuracy and run 16 times faster.
Tabular foundation models (TFMs) generate predictions on structured data via in-context learning, without task-specific estimation. We ask whether TFMs can be effectively applied to discrete choice, a central demand estimation framework in marketing and operations, and find that directly applying TFMs yields limited performance. The gap is structural: TFMs assume row-independent observations, whereas discrete choice is inherently set-valued and subject to persistent consumer preference heterogeneity. We propose a reformulation that encodes both choice-set dependence and individual heterogeneity within a row-based learning framework. Evaluated on a yogurt scanner panel, individual-level heterogeneity encoding is the dominant driver of predictive accuracy. The best reformulation outperforms hierarchical Bayesian estimation by 8\% in holdout log-likelihood and 3.6\% in hit rate, running 16 times faster, a practical advantage for large-scale demand estimation. The advantage is largest in the medium-data regime (10--40 purchase occasions per consumer), where parametric Bayesian shrinkage most distorts estimates for atypical consumers. Fine-tuning on population choice data provides additional gains for consumers with shallow purchase histories, where in-context learning has limited individual-specific signal to condition on. These results establish a principled approach for applying foundation models to consumer choice problems more broadly.
