Paper: arXiv 2611.13550

Authors: Ed deHaan, Allen H. Huang

Abstract

This technical note introduces tabular foundation models (TFMs) to accounting research and evaluates their performance in one common benchmark task: predicting firms’ next-year earnings changes. TFMs produce numeric predictions and learn in context, without post-release training or tuning, making them natural and easy to use for forecasting. The TFMs we study are pretrained on synthetic tabular data, so they avoid look-ahead bias from memorizing firms’ outcomes. We compare Google Research’s TabFM with a fine-tuned open-weight large language model developed in recent accounting research and with several tree-based models. TabFM is more accurate than all comparison models. It is also faster and less computationally intensive than the large language model. Several tree-based specifications produce predictions more quickly than TabFM, so choosing between them involves a trade-off between accuracy and speed. We conclude that TFMs have promise as a new tool for forecasting in accounting and finance. However, our evaluation is based on a single benchmark task, and no class of models is likely to dominate in all settings.

Complexity vs Empirical Score

  • Math Complexity: 5.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper introduces a novel application of Tabular Foundation Models to financial predictions with strong empirical validation. It thoroughly compares performance against established benchmarks and addresses potential biases, demonstrating robust methodology.

Research Flowchart

  flowchart TD
    A[Research Goal: Evaluate TFMs for Financial Predictions] --> B{Key Methodology: Compare models on next-year earnings changes};
    B --> C[Data/Inputs: Synthetic Tabular Data for TabFM, Historical Financial Data for Benchmarks];
    C --> D{Computational Processes: Contextual Learning (TFMs), Fine-tuning (LLM), Tree-based Model Training};
    D --> E[Models Evaluated: TabFM, Fine-tuned LLM, Tree-based Models];
    E --> F[Key Finding 1: TabFM is more accurate than all comparison models];
    F --> G[Key Finding 2: TabFM is faster and less computationally intensive than LLM];
    G --> H[Key Finding 3: Trade-off between accuracy (TabFM) and speed (some tree-based models)];
    H --> I[Conclusion: TFMs show promise for forecasting in accounting/finance, but more research is needed across tasks];